instruction stringlengths 34 186 | input stringlengths 2.02k 93.8k | output stringlengths 2 418 | meta_questiontype stringclasses 6 values | meta_inputlicense stringclasses 6 values | meta_pmid stringlengths 8 8 | meta_safetyreportid int64 9.51M 21M | meta_articlepubdate stringlengths 4 10 |
|---|---|---|---|---|---|---|---|
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Partial seizures'. | Focal bilateral motor seizures precipitated by abrupt cessation of chronic lormetazepam abuse and amitriptyline overdose.
We report the case of an adult psychiatric patient who developed new-onset focal bilateral motor seizures (FBMS) in the context of a severe benzodiazepine withdrawal syndrome. The patient was forced to interrupt chronic lormetazepam abuse and overdosed on amitriptyline (800 mg in an oral solution) before the onset of seizures. Typical signs of amitriptyline intoxication such as sedation and anticholinergic effects were not observed. Video-EEG recordings revealed a stereotypical ictal motor pattern with asymmetric tonic posturing and bilateral clonic movements of the upper limbs, but there were no abnormalities identified by EEG. Seizures recurred multiple times per day but resolved simultaneously when withdrawal symptomatology subsided eight days after onset. Nonepileptic seizures (NES) were considered in the differential diagnosis because of the patient's psychiatric history including preserved awareness during the bilateral convulsions, the absence of postictal confusion, and normal EEG. The present case indicates that FBMS may occur during benzodiazepine withdrawal in patients who overdosed on amitriptyline. The diagnosis may be challenging as FBMS may mimic NES in the absence of abnormal neurophysiologic findings. This may be especially challenging in patients with an underlying psychiatric disease.
1 Introduction
Epileptic seizures (ES) have been frequently reported as clinical manifestations of benzodiazepine withdrawal, an acute disorder precipitated by an abrupt discontinuation of chronic therapies including benzodiazepines. The withdrawal symptomatology may also include insomnia, irritability, muscular pain and stiffness, hand tremor, anxiety, and psychotic episodes [1]. An acute reduction in brain gamma-aminobutyric acid function has been indicated as a pathogenic mechanism of the syndrome [2]. While generalized tonic–clonic seizures are the most common, focal seizures are rarely reported in this condition. Particularly, focal nonconvulsive seizures have been described in patients treated with flumazenil during detoxification from lormetazepam abuse [3].
2 Case report
A 46-year-old woman with a clinical history of schizoaffective and personality disorders presented to our hospital with new-onset convulsive seizures. It was reported that the patient was forced to interrupt chronic lormetazepam abuse (a daily drug consumption of 150–250 mg of an oral solution) the day before the onset of seizures. When she ran out of the medication, she ingested an oral solution containing 800 mg of amitriptyline as a supposed sedative substitute for lormetazepam. Convulsive seizures recurred multiple times per day in the hospital during daytime and night-time hours in a clinical context of insomnia, restlessness, muscular rigidity, and painful spasms affecting the back muscles that implicated a benzodiazepine withdrawal syndrome. There was no evidence of amitriptyline toxicity, as the patient did not show sedation, anticholinergic effects, arterial hypotension, or cardiac arrhythmias. The amitriptyline blood level was not obtained. Seizures presented with a duration of approximately 53 s and a paroxysmal, stereotyped sequence of movements that was documented with video-EEG. The recordings of seizures, using the usual 10–20 system of electrode placement, did not show EEG abnormalities or postictal slowing of the background activity, but the tracings were disturbed by myogenic artifacts throughout the course of the events (Fig. 1, Fig. 2). Awareness was preserved during the episodes. The patient occasionally called for help at seizure onset but remained unresponsive until the cessation of movements. Rapid recovery of speech and an absence of postictal confusion were reported on different occasions. Magnetic resonance imaging of the brain was normal. Three days after the patient's admission, a cluster of convulsive episodes prompted a short intensive treatment with propofol and mechanical ventilation. Seizures were increasingly observed again after extubation, despite oral treatment with clonazepam, levetiracetam, and repeated intravenous administrations of diazepam. Regardless of the inefficacy of the treatment during an eight-day period of intense seizure activity, the patient eventually became seizure-free. The remission of seizures occurred simultaneously with the cessation of insomnia and muscular symptoms related to lormetazepam withdrawal. Clonazepam and levetiracetam were subsequently reduced and discontinued as the patient remained asymptomatic. Despite the absence of ictal EEG abnormalities, a retrospective analysis of the seizure semiology based on video recordings and additional recorded observations of the episodes at bedside suggested the clinical diagnosis of FBMS, presumably arising from the supplementary sensorimotor area (SSMA). This diagnostic conclusion was supported by the evidence of a paroxysmal, stereotyped ictal motor pattern with early abduction of the arm flexed at the elbow and subsequent asymmetric tonic posturing of the upper limb that reproduced the figure four sign [4] at seizure onset. The left arm was extended, and the right arm was flexed (at the elbow), while the head was briefly turned to the left at this point of the seizure. Consequent bilateral clonic movements of the upper limbs showed a typical epileptic pattern with a gradual frequency decline and a progressive amplitude increase of movements [5]. A stereotyped left hemifacial spasm was also noted during the tonic contraction of the upper limbs. The short duration of the episodes, occasional evidence of cyanosis and lack of pupillary reactivity served as additional clues to the ES diagnosis. Although the clinical picture was initially confounded by the patient's preserved awareness during the bilateral convulsions, the absence of postictal confusion, and the normal EEG findings, these features finally appeared consistent with FBMS arising in the SSMA, as previously reported in the literature [[6], [7], [8]]. The patient was discharged with a normal neurologic status nine days after the last convulsive episode. Seizures did not recur even without antiseizure medication during the following month, after which the patient was lost to follow-up.Fig. 1 EEG recording of the initial phase of a focal bilateral motor seizure with a still-frame from the video.
EEG recording of the initial phase of a focal bilateral motor seizure with a still-frame from the video.
The EEG does not reveal evident abnormalities at seizure onset. The tracing is progressively disturbed by myogenic artifacts during the evolution of the seizure motor pattern.
Fig. 1Fig. 2 EEG recording of the final phase of a focal bilateral motor seizure.
Phasic myogenic artifacts mark the tracing during the ending clonic phase of the seizure. The EEG shows the persistence of the background activity in the immediate postictal period.
Fig. 2
3 Discussion
The manifestation of convulsive seizures without EEG abnormalities in the context of benzodiazepine withdrawal syndrome and psychiatric disease requires a differential diagnosis of ES and nonepileptic seizures (NES). In fact, the abrupt interruption of high dosage, chronic benzodiazepine treatment represents a potential precipitating factor for ES, while NES are occasionally observed as imitators of epilepsy in patients with mental health problems [9,10]. In addition to the psychiatric disorders and the normal neurophysiologic findings, the patient's preserved awareness during the bilateral convulsions and the absence of postictal confusion confounded the clinical picture with the potential to misdiagnose FBMS as NES. FBMS are described as motor onset seizures arising in one cerebral hemisphere and rapidly involving bilateral motor networks, often with asymmetric tonic posturing [11]. The SSMA is classically recognized as the anatomical origin of FBMS [11]. Clinical characteristics of focal onset seizures arising in the SSMA include short duration (typically not exceeding 40 s) [6,12]; predominant onset during sleep [6,8]; somatosensory aura [8,13]; preserved awareness [6,8,13]; absence of postictal confusion [8]; vocalization or speech arrest [6,8,13]; unilateral tonic posturing [8,13] or bilateral, asymmetric tonic posturing [4,13,14]; bilateral abduction of the upper extremities [6]; and bilateral thrashing movements [6,8] that usually constitute a paroxysmal, stereotyped motor pattern in the same individual [6]. Particularly, asymmetric tonic posturing has been related to the asynchronous activation of the SSMA [4,14], thus representing a key semiologic feature of seizures involving this cortical area [8,12]. Authors have also referred to asymmetric tonic posturing as the figure four sign that marks focal to bilateral tonic–clonic seizures [4]. The lateralizing value of this sign has been repeatedly reported with the epileptogenic focus usually contralateral to the arm that is extended at the elbow [4,14]. In contrast, NES with convulsive-like manifestations are characterized by long duration (frequently exceeding 2 min); fluctuating course; and asynchronous movements (typically side-to-side head movements and out of phase limb movements) that show unchanged frequency and variable amplitude throughout the event [10].
In our patient, the convulsive disorders shared most of the mentioned clinical characteristics of focal onset seizures arising in the SSMA. In particular, the origin of ictal symptoms in the SSMA was suggested by the early stereotyped contraction of the left arm and consequent asymmetric tonic posturing of limbs, while a left hemifacial spasm appeared consistent with the presumed epileptic focus in the right cerebral hemisphere. The progression of these phenomena to bilateral, tonic extension and clonic movements of the upper extremities characterized the final phase of the motor pattern mimicking focal to bilateral tonic–clonic seizures. However, the absence of awareness impairments, postictal confusion and EEG slowing in the immediate postictal phase discredited the apparent bilateral tonic–clonic evolution of focal seizures, suggesting limited epileptic involvement of the SSMA during the entire course of the events. While the seizure semiology could be referred to FBMS arising from the SSMA, the absence of ictal EEG abnormalities to confirm epileptic seizures is a limiting factor for the diagnosis of frontal lobe seizures. This suggests the potential utility of EEG recordings with additional midline electrodes (placed according to the 10–10 system) and/or ictal SPECT, as a supplementary method to confirm the epileptic nature and the anatomical origin of the seizures. However, it is not uncommon for seizures arising in the mesial and deep structures of the frontal lobe to have absent scalp ictal EEG changes utilizing the 10-20 international system of electrode placement.
Although structural brain abnormalities were not documented in our patient, neurochemical alterations induced by benzodiazepine withdrawal into specific networks may have resulted in a major seizure susceptibility of the SSMA. Interestingly, a major intrinsic epileptogenicity of specific cortical areas has been postulated by Albiero et al. to explain the occurrence of focal seizures in patients who underwent detoxification from chronic lormetazepam abuse [3]. Additional information is needed to ascertain the major occurrence of focal seizures in patients who interrupted chronic lormetazepam abuse. Although amitriptyline may have lowered the seizure threshold, the precise clinical impact of the substitution remains uncertain in our patient, as typical symptoms of amitriptyline intoxication such as sedation, arterial hypotension, cardiac arrhythmias or anticholinergic effects [15] were not observed during the period of recurrent seizures. In contrast, the strict concomitance of seizures with classic benzodiazepine abstinence symptoms such as insomnia, restlessness, rigidity, and muscular pain clearly indicated the primary precipitating role of lormetazepam withdrawal. A previous report remarked that lormetazepam is the benzodiazepine that is most frequently associated with abuse and dependence [16], particularly when it is used in the form of an oral solution [17]. The presence of ethanol in the oral solution of lormetazepam has been specifically suspected to increase the risks for abuse and dependence. Although ethanol quantities may not be relevant in patients treated with accurate doses of lormetazepam (1.5–2.5 mg daily for insomnia), the pathogenic effect of ethanol may be significant in lormetazepam addicted patients. The present case provides additional information about the risks of lormetazepam addiction as a component of psychiatric care. We suggest strict monitoring of patients treated with an oral solution of this medication.
4 Conclusion
FBMS may be observed during the clinical manifestation of lormetazepam withdrawal in patients who overdosed on amitriptyline. FBMS may produce a confusing clinical picture, with the potential to misdiagnose NES, especially when no abnormal neurophysiologic findings are clearly identified and when patients have a known underlying psychiatric disease.
The following is the supplementary data related to this article.Video 1
Focal bilateral motor seizure precipitated by lormetazepam withdrawal and amitriptyline overdose.
The seizure begins with a tonic posture of the left hand and abduction of the left arm flexed at the elbow. The left arm is then elevated and extended anteriorly, while the right arm is initially flexed at the elbow and rotated inward, reproducing the figure four sign. The head is briefly turned to the left during the asymmetric posturing of the upper extremities. Consequently, the right arm is also extended anteriorly. A slight facial cyanosis and a grimace with left hemifacial spasm (consistent with the presumed origin of the seizure in the right cerebral hemisphere) are noted during the extension of the upper limbs. There is no eye blink at the threatening gesture or pupillary response to the light at this point of the seizure. Convulsions cease with a sequence of vibratory and clonic movements of the upper extremities and trunk. The duration of the seizure is approximately 53 s and amitriptyline overdose.
Video 1
Ethical statement
The authors state that the study described in the paper complies with the publishing ethics of the journal.
The patient’s face has been partially obscured in the attached video, as requested by the patient herself at the time of the given consent.
Declaration of competing interest
The authors state that there is no conflict of interest. | AMITRIPTYLINE, CLONAZEPAM, DIAZEPAM, LEVETIRACETAM, LORMETAZEPAM | DrugsGivenReaction | CC BY-NC-ND | 33490945 | 18,821,337 | 2021 |
What was the administration route of drug 'AMITRIPTYLINE'? | Focal bilateral motor seizures precipitated by abrupt cessation of chronic lormetazepam abuse and amitriptyline overdose.
We report the case of an adult psychiatric patient who developed new-onset focal bilateral motor seizures (FBMS) in the context of a severe benzodiazepine withdrawal syndrome. The patient was forced to interrupt chronic lormetazepam abuse and overdosed on amitriptyline (800 mg in an oral solution) before the onset of seizures. Typical signs of amitriptyline intoxication such as sedation and anticholinergic effects were not observed. Video-EEG recordings revealed a stereotypical ictal motor pattern with asymmetric tonic posturing and bilateral clonic movements of the upper limbs, but there were no abnormalities identified by EEG. Seizures recurred multiple times per day but resolved simultaneously when withdrawal symptomatology subsided eight days after onset. Nonepileptic seizures (NES) were considered in the differential diagnosis because of the patient's psychiatric history including preserved awareness during the bilateral convulsions, the absence of postictal confusion, and normal EEG. The present case indicates that FBMS may occur during benzodiazepine withdrawal in patients who overdosed on amitriptyline. The diagnosis may be challenging as FBMS may mimic NES in the absence of abnormal neurophysiologic findings. This may be especially challenging in patients with an underlying psychiatric disease.
1 Introduction
Epileptic seizures (ES) have been frequently reported as clinical manifestations of benzodiazepine withdrawal, an acute disorder precipitated by an abrupt discontinuation of chronic therapies including benzodiazepines. The withdrawal symptomatology may also include insomnia, irritability, muscular pain and stiffness, hand tremor, anxiety, and psychotic episodes [1]. An acute reduction in brain gamma-aminobutyric acid function has been indicated as a pathogenic mechanism of the syndrome [2]. While generalized tonic–clonic seizures are the most common, focal seizures are rarely reported in this condition. Particularly, focal nonconvulsive seizures have been described in patients treated with flumazenil during detoxification from lormetazepam abuse [3].
2 Case report
A 46-year-old woman with a clinical history of schizoaffective and personality disorders presented to our hospital with new-onset convulsive seizures. It was reported that the patient was forced to interrupt chronic lormetazepam abuse (a daily drug consumption of 150–250 mg of an oral solution) the day before the onset of seizures. When she ran out of the medication, she ingested an oral solution containing 800 mg of amitriptyline as a supposed sedative substitute for lormetazepam. Convulsive seizures recurred multiple times per day in the hospital during daytime and night-time hours in a clinical context of insomnia, restlessness, muscular rigidity, and painful spasms affecting the back muscles that implicated a benzodiazepine withdrawal syndrome. There was no evidence of amitriptyline toxicity, as the patient did not show sedation, anticholinergic effects, arterial hypotension, or cardiac arrhythmias. The amitriptyline blood level was not obtained. Seizures presented with a duration of approximately 53 s and a paroxysmal, stereotyped sequence of movements that was documented with video-EEG. The recordings of seizures, using the usual 10–20 system of electrode placement, did not show EEG abnormalities or postictal slowing of the background activity, but the tracings were disturbed by myogenic artifacts throughout the course of the events (Fig. 1, Fig. 2). Awareness was preserved during the episodes. The patient occasionally called for help at seizure onset but remained unresponsive until the cessation of movements. Rapid recovery of speech and an absence of postictal confusion were reported on different occasions. Magnetic resonance imaging of the brain was normal. Three days after the patient's admission, a cluster of convulsive episodes prompted a short intensive treatment with propofol and mechanical ventilation. Seizures were increasingly observed again after extubation, despite oral treatment with clonazepam, levetiracetam, and repeated intravenous administrations of diazepam. Regardless of the inefficacy of the treatment during an eight-day period of intense seizure activity, the patient eventually became seizure-free. The remission of seizures occurred simultaneously with the cessation of insomnia and muscular symptoms related to lormetazepam withdrawal. Clonazepam and levetiracetam were subsequently reduced and discontinued as the patient remained asymptomatic. Despite the absence of ictal EEG abnormalities, a retrospective analysis of the seizure semiology based on video recordings and additional recorded observations of the episodes at bedside suggested the clinical diagnosis of FBMS, presumably arising from the supplementary sensorimotor area (SSMA). This diagnostic conclusion was supported by the evidence of a paroxysmal, stereotyped ictal motor pattern with early abduction of the arm flexed at the elbow and subsequent asymmetric tonic posturing of the upper limb that reproduced the figure four sign [4] at seizure onset. The left arm was extended, and the right arm was flexed (at the elbow), while the head was briefly turned to the left at this point of the seizure. Consequent bilateral clonic movements of the upper limbs showed a typical epileptic pattern with a gradual frequency decline and a progressive amplitude increase of movements [5]. A stereotyped left hemifacial spasm was also noted during the tonic contraction of the upper limbs. The short duration of the episodes, occasional evidence of cyanosis and lack of pupillary reactivity served as additional clues to the ES diagnosis. Although the clinical picture was initially confounded by the patient's preserved awareness during the bilateral convulsions, the absence of postictal confusion, and the normal EEG findings, these features finally appeared consistent with FBMS arising in the SSMA, as previously reported in the literature [[6], [7], [8]]. The patient was discharged with a normal neurologic status nine days after the last convulsive episode. Seizures did not recur even without antiseizure medication during the following month, after which the patient was lost to follow-up.Fig. 1 EEG recording of the initial phase of a focal bilateral motor seizure with a still-frame from the video.
EEG recording of the initial phase of a focal bilateral motor seizure with a still-frame from the video.
The EEG does not reveal evident abnormalities at seizure onset. The tracing is progressively disturbed by myogenic artifacts during the evolution of the seizure motor pattern.
Fig. 1Fig. 2 EEG recording of the final phase of a focal bilateral motor seizure.
Phasic myogenic artifacts mark the tracing during the ending clonic phase of the seizure. The EEG shows the persistence of the background activity in the immediate postictal period.
Fig. 2
3 Discussion
The manifestation of convulsive seizures without EEG abnormalities in the context of benzodiazepine withdrawal syndrome and psychiatric disease requires a differential diagnosis of ES and nonepileptic seizures (NES). In fact, the abrupt interruption of high dosage, chronic benzodiazepine treatment represents a potential precipitating factor for ES, while NES are occasionally observed as imitators of epilepsy in patients with mental health problems [9,10]. In addition to the psychiatric disorders and the normal neurophysiologic findings, the patient's preserved awareness during the bilateral convulsions and the absence of postictal confusion confounded the clinical picture with the potential to misdiagnose FBMS as NES. FBMS are described as motor onset seizures arising in one cerebral hemisphere and rapidly involving bilateral motor networks, often with asymmetric tonic posturing [11]. The SSMA is classically recognized as the anatomical origin of FBMS [11]. Clinical characteristics of focal onset seizures arising in the SSMA include short duration (typically not exceeding 40 s) [6,12]; predominant onset during sleep [6,8]; somatosensory aura [8,13]; preserved awareness [6,8,13]; absence of postictal confusion [8]; vocalization or speech arrest [6,8,13]; unilateral tonic posturing [8,13] or bilateral, asymmetric tonic posturing [4,13,14]; bilateral abduction of the upper extremities [6]; and bilateral thrashing movements [6,8] that usually constitute a paroxysmal, stereotyped motor pattern in the same individual [6]. Particularly, asymmetric tonic posturing has been related to the asynchronous activation of the SSMA [4,14], thus representing a key semiologic feature of seizures involving this cortical area [8,12]. Authors have also referred to asymmetric tonic posturing as the figure four sign that marks focal to bilateral tonic–clonic seizures [4]. The lateralizing value of this sign has been repeatedly reported with the epileptogenic focus usually contralateral to the arm that is extended at the elbow [4,14]. In contrast, NES with convulsive-like manifestations are characterized by long duration (frequently exceeding 2 min); fluctuating course; and asynchronous movements (typically side-to-side head movements and out of phase limb movements) that show unchanged frequency and variable amplitude throughout the event [10].
In our patient, the convulsive disorders shared most of the mentioned clinical characteristics of focal onset seizures arising in the SSMA. In particular, the origin of ictal symptoms in the SSMA was suggested by the early stereotyped contraction of the left arm and consequent asymmetric tonic posturing of limbs, while a left hemifacial spasm appeared consistent with the presumed epileptic focus in the right cerebral hemisphere. The progression of these phenomena to bilateral, tonic extension and clonic movements of the upper extremities characterized the final phase of the motor pattern mimicking focal to bilateral tonic–clonic seizures. However, the absence of awareness impairments, postictal confusion and EEG slowing in the immediate postictal phase discredited the apparent bilateral tonic–clonic evolution of focal seizures, suggesting limited epileptic involvement of the SSMA during the entire course of the events. While the seizure semiology could be referred to FBMS arising from the SSMA, the absence of ictal EEG abnormalities to confirm epileptic seizures is a limiting factor for the diagnosis of frontal lobe seizures. This suggests the potential utility of EEG recordings with additional midline electrodes (placed according to the 10–10 system) and/or ictal SPECT, as a supplementary method to confirm the epileptic nature and the anatomical origin of the seizures. However, it is not uncommon for seizures arising in the mesial and deep structures of the frontal lobe to have absent scalp ictal EEG changes utilizing the 10-20 international system of electrode placement.
Although structural brain abnormalities were not documented in our patient, neurochemical alterations induced by benzodiazepine withdrawal into specific networks may have resulted in a major seizure susceptibility of the SSMA. Interestingly, a major intrinsic epileptogenicity of specific cortical areas has been postulated by Albiero et al. to explain the occurrence of focal seizures in patients who underwent detoxification from chronic lormetazepam abuse [3]. Additional information is needed to ascertain the major occurrence of focal seizures in patients who interrupted chronic lormetazepam abuse. Although amitriptyline may have lowered the seizure threshold, the precise clinical impact of the substitution remains uncertain in our patient, as typical symptoms of amitriptyline intoxication such as sedation, arterial hypotension, cardiac arrhythmias or anticholinergic effects [15] were not observed during the period of recurrent seizures. In contrast, the strict concomitance of seizures with classic benzodiazepine abstinence symptoms such as insomnia, restlessness, rigidity, and muscular pain clearly indicated the primary precipitating role of lormetazepam withdrawal. A previous report remarked that lormetazepam is the benzodiazepine that is most frequently associated with abuse and dependence [16], particularly when it is used in the form of an oral solution [17]. The presence of ethanol in the oral solution of lormetazepam has been specifically suspected to increase the risks for abuse and dependence. Although ethanol quantities may not be relevant in patients treated with accurate doses of lormetazepam (1.5–2.5 mg daily for insomnia), the pathogenic effect of ethanol may be significant in lormetazepam addicted patients. The present case provides additional information about the risks of lormetazepam addiction as a component of psychiatric care. We suggest strict monitoring of patients treated with an oral solution of this medication.
4 Conclusion
FBMS may be observed during the clinical manifestation of lormetazepam withdrawal in patients who overdosed on amitriptyline. FBMS may produce a confusing clinical picture, with the potential to misdiagnose NES, especially when no abnormal neurophysiologic findings are clearly identified and when patients have a known underlying psychiatric disease.
The following is the supplementary data related to this article.Video 1
Focal bilateral motor seizure precipitated by lormetazepam withdrawal and amitriptyline overdose.
The seizure begins with a tonic posture of the left hand and abduction of the left arm flexed at the elbow. The left arm is then elevated and extended anteriorly, while the right arm is initially flexed at the elbow and rotated inward, reproducing the figure four sign. The head is briefly turned to the left during the asymmetric posturing of the upper extremities. Consequently, the right arm is also extended anteriorly. A slight facial cyanosis and a grimace with left hemifacial spasm (consistent with the presumed origin of the seizure in the right cerebral hemisphere) are noted during the extension of the upper limbs. There is no eye blink at the threatening gesture or pupillary response to the light at this point of the seizure. Convulsions cease with a sequence of vibratory and clonic movements of the upper extremities and trunk. The duration of the seizure is approximately 53 s and amitriptyline overdose.
Video 1
Ethical statement
The authors state that the study described in the paper complies with the publishing ethics of the journal.
The patient’s face has been partially obscured in the attached video, as requested by the patient herself at the time of the given consent.
Declaration of competing interest
The authors state that there is no conflict of interest. | Oral | DrugAdministrationRoute | CC BY-NC-ND | 33490945 | 18,821,337 | 2021 |
What was the administration route of drug 'CLONAZEPAM'? | Focal bilateral motor seizures precipitated by abrupt cessation of chronic lormetazepam abuse and amitriptyline overdose.
We report the case of an adult psychiatric patient who developed new-onset focal bilateral motor seizures (FBMS) in the context of a severe benzodiazepine withdrawal syndrome. The patient was forced to interrupt chronic lormetazepam abuse and overdosed on amitriptyline (800 mg in an oral solution) before the onset of seizures. Typical signs of amitriptyline intoxication such as sedation and anticholinergic effects were not observed. Video-EEG recordings revealed a stereotypical ictal motor pattern with asymmetric tonic posturing and bilateral clonic movements of the upper limbs, but there were no abnormalities identified by EEG. Seizures recurred multiple times per day but resolved simultaneously when withdrawal symptomatology subsided eight days after onset. Nonepileptic seizures (NES) were considered in the differential diagnosis because of the patient's psychiatric history including preserved awareness during the bilateral convulsions, the absence of postictal confusion, and normal EEG. The present case indicates that FBMS may occur during benzodiazepine withdrawal in patients who overdosed on amitriptyline. The diagnosis may be challenging as FBMS may mimic NES in the absence of abnormal neurophysiologic findings. This may be especially challenging in patients with an underlying psychiatric disease.
1 Introduction
Epileptic seizures (ES) have been frequently reported as clinical manifestations of benzodiazepine withdrawal, an acute disorder precipitated by an abrupt discontinuation of chronic therapies including benzodiazepines. The withdrawal symptomatology may also include insomnia, irritability, muscular pain and stiffness, hand tremor, anxiety, and psychotic episodes [1]. An acute reduction in brain gamma-aminobutyric acid function has been indicated as a pathogenic mechanism of the syndrome [2]. While generalized tonic–clonic seizures are the most common, focal seizures are rarely reported in this condition. Particularly, focal nonconvulsive seizures have been described in patients treated with flumazenil during detoxification from lormetazepam abuse [3].
2 Case report
A 46-year-old woman with a clinical history of schizoaffective and personality disorders presented to our hospital with new-onset convulsive seizures. It was reported that the patient was forced to interrupt chronic lormetazepam abuse (a daily drug consumption of 150–250 mg of an oral solution) the day before the onset of seizures. When she ran out of the medication, she ingested an oral solution containing 800 mg of amitriptyline as a supposed sedative substitute for lormetazepam. Convulsive seizures recurred multiple times per day in the hospital during daytime and night-time hours in a clinical context of insomnia, restlessness, muscular rigidity, and painful spasms affecting the back muscles that implicated a benzodiazepine withdrawal syndrome. There was no evidence of amitriptyline toxicity, as the patient did not show sedation, anticholinergic effects, arterial hypotension, or cardiac arrhythmias. The amitriptyline blood level was not obtained. Seizures presented with a duration of approximately 53 s and a paroxysmal, stereotyped sequence of movements that was documented with video-EEG. The recordings of seizures, using the usual 10–20 system of electrode placement, did not show EEG abnormalities or postictal slowing of the background activity, but the tracings were disturbed by myogenic artifacts throughout the course of the events (Fig. 1, Fig. 2). Awareness was preserved during the episodes. The patient occasionally called for help at seizure onset but remained unresponsive until the cessation of movements. Rapid recovery of speech and an absence of postictal confusion were reported on different occasions. Magnetic resonance imaging of the brain was normal. Three days after the patient's admission, a cluster of convulsive episodes prompted a short intensive treatment with propofol and mechanical ventilation. Seizures were increasingly observed again after extubation, despite oral treatment with clonazepam, levetiracetam, and repeated intravenous administrations of diazepam. Regardless of the inefficacy of the treatment during an eight-day period of intense seizure activity, the patient eventually became seizure-free. The remission of seizures occurred simultaneously with the cessation of insomnia and muscular symptoms related to lormetazepam withdrawal. Clonazepam and levetiracetam were subsequently reduced and discontinued as the patient remained asymptomatic. Despite the absence of ictal EEG abnormalities, a retrospective analysis of the seizure semiology based on video recordings and additional recorded observations of the episodes at bedside suggested the clinical diagnosis of FBMS, presumably arising from the supplementary sensorimotor area (SSMA). This diagnostic conclusion was supported by the evidence of a paroxysmal, stereotyped ictal motor pattern with early abduction of the arm flexed at the elbow and subsequent asymmetric tonic posturing of the upper limb that reproduced the figure four sign [4] at seizure onset. The left arm was extended, and the right arm was flexed (at the elbow), while the head was briefly turned to the left at this point of the seizure. Consequent bilateral clonic movements of the upper limbs showed a typical epileptic pattern with a gradual frequency decline and a progressive amplitude increase of movements [5]. A stereotyped left hemifacial spasm was also noted during the tonic contraction of the upper limbs. The short duration of the episodes, occasional evidence of cyanosis and lack of pupillary reactivity served as additional clues to the ES diagnosis. Although the clinical picture was initially confounded by the patient's preserved awareness during the bilateral convulsions, the absence of postictal confusion, and the normal EEG findings, these features finally appeared consistent with FBMS arising in the SSMA, as previously reported in the literature [[6], [7], [8]]. The patient was discharged with a normal neurologic status nine days after the last convulsive episode. Seizures did not recur even without antiseizure medication during the following month, after which the patient was lost to follow-up.Fig. 1 EEG recording of the initial phase of a focal bilateral motor seizure with a still-frame from the video.
EEG recording of the initial phase of a focal bilateral motor seizure with a still-frame from the video.
The EEG does not reveal evident abnormalities at seizure onset. The tracing is progressively disturbed by myogenic artifacts during the evolution of the seizure motor pattern.
Fig. 1Fig. 2 EEG recording of the final phase of a focal bilateral motor seizure.
Phasic myogenic artifacts mark the tracing during the ending clonic phase of the seizure. The EEG shows the persistence of the background activity in the immediate postictal period.
Fig. 2
3 Discussion
The manifestation of convulsive seizures without EEG abnormalities in the context of benzodiazepine withdrawal syndrome and psychiatric disease requires a differential diagnosis of ES and nonepileptic seizures (NES). In fact, the abrupt interruption of high dosage, chronic benzodiazepine treatment represents a potential precipitating factor for ES, while NES are occasionally observed as imitators of epilepsy in patients with mental health problems [9,10]. In addition to the psychiatric disorders and the normal neurophysiologic findings, the patient's preserved awareness during the bilateral convulsions and the absence of postictal confusion confounded the clinical picture with the potential to misdiagnose FBMS as NES. FBMS are described as motor onset seizures arising in one cerebral hemisphere and rapidly involving bilateral motor networks, often with asymmetric tonic posturing [11]. The SSMA is classically recognized as the anatomical origin of FBMS [11]. Clinical characteristics of focal onset seizures arising in the SSMA include short duration (typically not exceeding 40 s) [6,12]; predominant onset during sleep [6,8]; somatosensory aura [8,13]; preserved awareness [6,8,13]; absence of postictal confusion [8]; vocalization or speech arrest [6,8,13]; unilateral tonic posturing [8,13] or bilateral, asymmetric tonic posturing [4,13,14]; bilateral abduction of the upper extremities [6]; and bilateral thrashing movements [6,8] that usually constitute a paroxysmal, stereotyped motor pattern in the same individual [6]. Particularly, asymmetric tonic posturing has been related to the asynchronous activation of the SSMA [4,14], thus representing a key semiologic feature of seizures involving this cortical area [8,12]. Authors have also referred to asymmetric tonic posturing as the figure four sign that marks focal to bilateral tonic–clonic seizures [4]. The lateralizing value of this sign has been repeatedly reported with the epileptogenic focus usually contralateral to the arm that is extended at the elbow [4,14]. In contrast, NES with convulsive-like manifestations are characterized by long duration (frequently exceeding 2 min); fluctuating course; and asynchronous movements (typically side-to-side head movements and out of phase limb movements) that show unchanged frequency and variable amplitude throughout the event [10].
In our patient, the convulsive disorders shared most of the mentioned clinical characteristics of focal onset seizures arising in the SSMA. In particular, the origin of ictal symptoms in the SSMA was suggested by the early stereotyped contraction of the left arm and consequent asymmetric tonic posturing of limbs, while a left hemifacial spasm appeared consistent with the presumed epileptic focus in the right cerebral hemisphere. The progression of these phenomena to bilateral, tonic extension and clonic movements of the upper extremities characterized the final phase of the motor pattern mimicking focal to bilateral tonic–clonic seizures. However, the absence of awareness impairments, postictal confusion and EEG slowing in the immediate postictal phase discredited the apparent bilateral tonic–clonic evolution of focal seizures, suggesting limited epileptic involvement of the SSMA during the entire course of the events. While the seizure semiology could be referred to FBMS arising from the SSMA, the absence of ictal EEG abnormalities to confirm epileptic seizures is a limiting factor for the diagnosis of frontal lobe seizures. This suggests the potential utility of EEG recordings with additional midline electrodes (placed according to the 10–10 system) and/or ictal SPECT, as a supplementary method to confirm the epileptic nature and the anatomical origin of the seizures. However, it is not uncommon for seizures arising in the mesial and deep structures of the frontal lobe to have absent scalp ictal EEG changes utilizing the 10-20 international system of electrode placement.
Although structural brain abnormalities were not documented in our patient, neurochemical alterations induced by benzodiazepine withdrawal into specific networks may have resulted in a major seizure susceptibility of the SSMA. Interestingly, a major intrinsic epileptogenicity of specific cortical areas has been postulated by Albiero et al. to explain the occurrence of focal seizures in patients who underwent detoxification from chronic lormetazepam abuse [3]. Additional information is needed to ascertain the major occurrence of focal seizures in patients who interrupted chronic lormetazepam abuse. Although amitriptyline may have lowered the seizure threshold, the precise clinical impact of the substitution remains uncertain in our patient, as typical symptoms of amitriptyline intoxication such as sedation, arterial hypotension, cardiac arrhythmias or anticholinergic effects [15] were not observed during the period of recurrent seizures. In contrast, the strict concomitance of seizures with classic benzodiazepine abstinence symptoms such as insomnia, restlessness, rigidity, and muscular pain clearly indicated the primary precipitating role of lormetazepam withdrawal. A previous report remarked that lormetazepam is the benzodiazepine that is most frequently associated with abuse and dependence [16], particularly when it is used in the form of an oral solution [17]. The presence of ethanol in the oral solution of lormetazepam has been specifically suspected to increase the risks for abuse and dependence. Although ethanol quantities may not be relevant in patients treated with accurate doses of lormetazepam (1.5–2.5 mg daily for insomnia), the pathogenic effect of ethanol may be significant in lormetazepam addicted patients. The present case provides additional information about the risks of lormetazepam addiction as a component of psychiatric care. We suggest strict monitoring of patients treated with an oral solution of this medication.
4 Conclusion
FBMS may be observed during the clinical manifestation of lormetazepam withdrawal in patients who overdosed on amitriptyline. FBMS may produce a confusing clinical picture, with the potential to misdiagnose NES, especially when no abnormal neurophysiologic findings are clearly identified and when patients have a known underlying psychiatric disease.
The following is the supplementary data related to this article.Video 1
Focal bilateral motor seizure precipitated by lormetazepam withdrawal and amitriptyline overdose.
The seizure begins with a tonic posture of the left hand and abduction of the left arm flexed at the elbow. The left arm is then elevated and extended anteriorly, while the right arm is initially flexed at the elbow and rotated inward, reproducing the figure four sign. The head is briefly turned to the left during the asymmetric posturing of the upper extremities. Consequently, the right arm is also extended anteriorly. A slight facial cyanosis and a grimace with left hemifacial spasm (consistent with the presumed origin of the seizure in the right cerebral hemisphere) are noted during the extension of the upper limbs. There is no eye blink at the threatening gesture or pupillary response to the light at this point of the seizure. Convulsions cease with a sequence of vibratory and clonic movements of the upper extremities and trunk. The duration of the seizure is approximately 53 s and amitriptyline overdose.
Video 1
Ethical statement
The authors state that the study described in the paper complies with the publishing ethics of the journal.
The patient’s face has been partially obscured in the attached video, as requested by the patient herself at the time of the given consent.
Declaration of competing interest
The authors state that there is no conflict of interest. | Oral | DrugAdministrationRoute | CC BY-NC-ND | 33490945 | 18,821,337 | 2021 |
What was the administration route of drug 'DIAZEPAM'? | Focal bilateral motor seizures precipitated by abrupt cessation of chronic lormetazepam abuse and amitriptyline overdose.
We report the case of an adult psychiatric patient who developed new-onset focal bilateral motor seizures (FBMS) in the context of a severe benzodiazepine withdrawal syndrome. The patient was forced to interrupt chronic lormetazepam abuse and overdosed on amitriptyline (800 mg in an oral solution) before the onset of seizures. Typical signs of amitriptyline intoxication such as sedation and anticholinergic effects were not observed. Video-EEG recordings revealed a stereotypical ictal motor pattern with asymmetric tonic posturing and bilateral clonic movements of the upper limbs, but there were no abnormalities identified by EEG. Seizures recurred multiple times per day but resolved simultaneously when withdrawal symptomatology subsided eight days after onset. Nonepileptic seizures (NES) were considered in the differential diagnosis because of the patient's psychiatric history including preserved awareness during the bilateral convulsions, the absence of postictal confusion, and normal EEG. The present case indicates that FBMS may occur during benzodiazepine withdrawal in patients who overdosed on amitriptyline. The diagnosis may be challenging as FBMS may mimic NES in the absence of abnormal neurophysiologic findings. This may be especially challenging in patients with an underlying psychiatric disease.
1 Introduction
Epileptic seizures (ES) have been frequently reported as clinical manifestations of benzodiazepine withdrawal, an acute disorder precipitated by an abrupt discontinuation of chronic therapies including benzodiazepines. The withdrawal symptomatology may also include insomnia, irritability, muscular pain and stiffness, hand tremor, anxiety, and psychotic episodes [1]. An acute reduction in brain gamma-aminobutyric acid function has been indicated as a pathogenic mechanism of the syndrome [2]. While generalized tonic–clonic seizures are the most common, focal seizures are rarely reported in this condition. Particularly, focal nonconvulsive seizures have been described in patients treated with flumazenil during detoxification from lormetazepam abuse [3].
2 Case report
A 46-year-old woman with a clinical history of schizoaffective and personality disorders presented to our hospital with new-onset convulsive seizures. It was reported that the patient was forced to interrupt chronic lormetazepam abuse (a daily drug consumption of 150–250 mg of an oral solution) the day before the onset of seizures. When she ran out of the medication, she ingested an oral solution containing 800 mg of amitriptyline as a supposed sedative substitute for lormetazepam. Convulsive seizures recurred multiple times per day in the hospital during daytime and night-time hours in a clinical context of insomnia, restlessness, muscular rigidity, and painful spasms affecting the back muscles that implicated a benzodiazepine withdrawal syndrome. There was no evidence of amitriptyline toxicity, as the patient did not show sedation, anticholinergic effects, arterial hypotension, or cardiac arrhythmias. The amitriptyline blood level was not obtained. Seizures presented with a duration of approximately 53 s and a paroxysmal, stereotyped sequence of movements that was documented with video-EEG. The recordings of seizures, using the usual 10–20 system of electrode placement, did not show EEG abnormalities or postictal slowing of the background activity, but the tracings were disturbed by myogenic artifacts throughout the course of the events (Fig. 1, Fig. 2). Awareness was preserved during the episodes. The patient occasionally called for help at seizure onset but remained unresponsive until the cessation of movements. Rapid recovery of speech and an absence of postictal confusion were reported on different occasions. Magnetic resonance imaging of the brain was normal. Three days after the patient's admission, a cluster of convulsive episodes prompted a short intensive treatment with propofol and mechanical ventilation. Seizures were increasingly observed again after extubation, despite oral treatment with clonazepam, levetiracetam, and repeated intravenous administrations of diazepam. Regardless of the inefficacy of the treatment during an eight-day period of intense seizure activity, the patient eventually became seizure-free. The remission of seizures occurred simultaneously with the cessation of insomnia and muscular symptoms related to lormetazepam withdrawal. Clonazepam and levetiracetam were subsequently reduced and discontinued as the patient remained asymptomatic. Despite the absence of ictal EEG abnormalities, a retrospective analysis of the seizure semiology based on video recordings and additional recorded observations of the episodes at bedside suggested the clinical diagnosis of FBMS, presumably arising from the supplementary sensorimotor area (SSMA). This diagnostic conclusion was supported by the evidence of a paroxysmal, stereotyped ictal motor pattern with early abduction of the arm flexed at the elbow and subsequent asymmetric tonic posturing of the upper limb that reproduced the figure four sign [4] at seizure onset. The left arm was extended, and the right arm was flexed (at the elbow), while the head was briefly turned to the left at this point of the seizure. Consequent bilateral clonic movements of the upper limbs showed a typical epileptic pattern with a gradual frequency decline and a progressive amplitude increase of movements [5]. A stereotyped left hemifacial spasm was also noted during the tonic contraction of the upper limbs. The short duration of the episodes, occasional evidence of cyanosis and lack of pupillary reactivity served as additional clues to the ES diagnosis. Although the clinical picture was initially confounded by the patient's preserved awareness during the bilateral convulsions, the absence of postictal confusion, and the normal EEG findings, these features finally appeared consistent with FBMS arising in the SSMA, as previously reported in the literature [[6], [7], [8]]. The patient was discharged with a normal neurologic status nine days after the last convulsive episode. Seizures did not recur even without antiseizure medication during the following month, after which the patient was lost to follow-up.Fig. 1 EEG recording of the initial phase of a focal bilateral motor seizure with a still-frame from the video.
EEG recording of the initial phase of a focal bilateral motor seizure with a still-frame from the video.
The EEG does not reveal evident abnormalities at seizure onset. The tracing is progressively disturbed by myogenic artifacts during the evolution of the seizure motor pattern.
Fig. 1Fig. 2 EEG recording of the final phase of a focal bilateral motor seizure.
Phasic myogenic artifacts mark the tracing during the ending clonic phase of the seizure. The EEG shows the persistence of the background activity in the immediate postictal period.
Fig. 2
3 Discussion
The manifestation of convulsive seizures without EEG abnormalities in the context of benzodiazepine withdrawal syndrome and psychiatric disease requires a differential diagnosis of ES and nonepileptic seizures (NES). In fact, the abrupt interruption of high dosage, chronic benzodiazepine treatment represents a potential precipitating factor for ES, while NES are occasionally observed as imitators of epilepsy in patients with mental health problems [9,10]. In addition to the psychiatric disorders and the normal neurophysiologic findings, the patient's preserved awareness during the bilateral convulsions and the absence of postictal confusion confounded the clinical picture with the potential to misdiagnose FBMS as NES. FBMS are described as motor onset seizures arising in one cerebral hemisphere and rapidly involving bilateral motor networks, often with asymmetric tonic posturing [11]. The SSMA is classically recognized as the anatomical origin of FBMS [11]. Clinical characteristics of focal onset seizures arising in the SSMA include short duration (typically not exceeding 40 s) [6,12]; predominant onset during sleep [6,8]; somatosensory aura [8,13]; preserved awareness [6,8,13]; absence of postictal confusion [8]; vocalization or speech arrest [6,8,13]; unilateral tonic posturing [8,13] or bilateral, asymmetric tonic posturing [4,13,14]; bilateral abduction of the upper extremities [6]; and bilateral thrashing movements [6,8] that usually constitute a paroxysmal, stereotyped motor pattern in the same individual [6]. Particularly, asymmetric tonic posturing has been related to the asynchronous activation of the SSMA [4,14], thus representing a key semiologic feature of seizures involving this cortical area [8,12]. Authors have also referred to asymmetric tonic posturing as the figure four sign that marks focal to bilateral tonic–clonic seizures [4]. The lateralizing value of this sign has been repeatedly reported with the epileptogenic focus usually contralateral to the arm that is extended at the elbow [4,14]. In contrast, NES with convulsive-like manifestations are characterized by long duration (frequently exceeding 2 min); fluctuating course; and asynchronous movements (typically side-to-side head movements and out of phase limb movements) that show unchanged frequency and variable amplitude throughout the event [10].
In our patient, the convulsive disorders shared most of the mentioned clinical characteristics of focal onset seizures arising in the SSMA. In particular, the origin of ictal symptoms in the SSMA was suggested by the early stereotyped contraction of the left arm and consequent asymmetric tonic posturing of limbs, while a left hemifacial spasm appeared consistent with the presumed epileptic focus in the right cerebral hemisphere. The progression of these phenomena to bilateral, tonic extension and clonic movements of the upper extremities characterized the final phase of the motor pattern mimicking focal to bilateral tonic–clonic seizures. However, the absence of awareness impairments, postictal confusion and EEG slowing in the immediate postictal phase discredited the apparent bilateral tonic–clonic evolution of focal seizures, suggesting limited epileptic involvement of the SSMA during the entire course of the events. While the seizure semiology could be referred to FBMS arising from the SSMA, the absence of ictal EEG abnormalities to confirm epileptic seizures is a limiting factor for the diagnosis of frontal lobe seizures. This suggests the potential utility of EEG recordings with additional midline electrodes (placed according to the 10–10 system) and/or ictal SPECT, as a supplementary method to confirm the epileptic nature and the anatomical origin of the seizures. However, it is not uncommon for seizures arising in the mesial and deep structures of the frontal lobe to have absent scalp ictal EEG changes utilizing the 10-20 international system of electrode placement.
Although structural brain abnormalities were not documented in our patient, neurochemical alterations induced by benzodiazepine withdrawal into specific networks may have resulted in a major seizure susceptibility of the SSMA. Interestingly, a major intrinsic epileptogenicity of specific cortical areas has been postulated by Albiero et al. to explain the occurrence of focal seizures in patients who underwent detoxification from chronic lormetazepam abuse [3]. Additional information is needed to ascertain the major occurrence of focal seizures in patients who interrupted chronic lormetazepam abuse. Although amitriptyline may have lowered the seizure threshold, the precise clinical impact of the substitution remains uncertain in our patient, as typical symptoms of amitriptyline intoxication such as sedation, arterial hypotension, cardiac arrhythmias or anticholinergic effects [15] were not observed during the period of recurrent seizures. In contrast, the strict concomitance of seizures with classic benzodiazepine abstinence symptoms such as insomnia, restlessness, rigidity, and muscular pain clearly indicated the primary precipitating role of lormetazepam withdrawal. A previous report remarked that lormetazepam is the benzodiazepine that is most frequently associated with abuse and dependence [16], particularly when it is used in the form of an oral solution [17]. The presence of ethanol in the oral solution of lormetazepam has been specifically suspected to increase the risks for abuse and dependence. Although ethanol quantities may not be relevant in patients treated with accurate doses of lormetazepam (1.5–2.5 mg daily for insomnia), the pathogenic effect of ethanol may be significant in lormetazepam addicted patients. The present case provides additional information about the risks of lormetazepam addiction as a component of psychiatric care. We suggest strict monitoring of patients treated with an oral solution of this medication.
4 Conclusion
FBMS may be observed during the clinical manifestation of lormetazepam withdrawal in patients who overdosed on amitriptyline. FBMS may produce a confusing clinical picture, with the potential to misdiagnose NES, especially when no abnormal neurophysiologic findings are clearly identified and when patients have a known underlying psychiatric disease.
The following is the supplementary data related to this article.Video 1
Focal bilateral motor seizure precipitated by lormetazepam withdrawal and amitriptyline overdose.
The seizure begins with a tonic posture of the left hand and abduction of the left arm flexed at the elbow. The left arm is then elevated and extended anteriorly, while the right arm is initially flexed at the elbow and rotated inward, reproducing the figure four sign. The head is briefly turned to the left during the asymmetric posturing of the upper extremities. Consequently, the right arm is also extended anteriorly. A slight facial cyanosis and a grimace with left hemifacial spasm (consistent with the presumed origin of the seizure in the right cerebral hemisphere) are noted during the extension of the upper limbs. There is no eye blink at the threatening gesture or pupillary response to the light at this point of the seizure. Convulsions cease with a sequence of vibratory and clonic movements of the upper extremities and trunk. The duration of the seizure is approximately 53 s and amitriptyline overdose.
Video 1
Ethical statement
The authors state that the study described in the paper complies with the publishing ethics of the journal.
The patient’s face has been partially obscured in the attached video, as requested by the patient herself at the time of the given consent.
Declaration of competing interest
The authors state that there is no conflict of interest. | Intravenous (not otherwise specified) | DrugAdministrationRoute | CC BY-NC-ND | 33490945 | 18,821,337 | 2021 |
What was the administration route of drug 'LEVETIRACETAM'? | Focal bilateral motor seizures precipitated by abrupt cessation of chronic lormetazepam abuse and amitriptyline overdose.
We report the case of an adult psychiatric patient who developed new-onset focal bilateral motor seizures (FBMS) in the context of a severe benzodiazepine withdrawal syndrome. The patient was forced to interrupt chronic lormetazepam abuse and overdosed on amitriptyline (800 mg in an oral solution) before the onset of seizures. Typical signs of amitriptyline intoxication such as sedation and anticholinergic effects were not observed. Video-EEG recordings revealed a stereotypical ictal motor pattern with asymmetric tonic posturing and bilateral clonic movements of the upper limbs, but there were no abnormalities identified by EEG. Seizures recurred multiple times per day but resolved simultaneously when withdrawal symptomatology subsided eight days after onset. Nonepileptic seizures (NES) were considered in the differential diagnosis because of the patient's psychiatric history including preserved awareness during the bilateral convulsions, the absence of postictal confusion, and normal EEG. The present case indicates that FBMS may occur during benzodiazepine withdrawal in patients who overdosed on amitriptyline. The diagnosis may be challenging as FBMS may mimic NES in the absence of abnormal neurophysiologic findings. This may be especially challenging in patients with an underlying psychiatric disease.
1 Introduction
Epileptic seizures (ES) have been frequently reported as clinical manifestations of benzodiazepine withdrawal, an acute disorder precipitated by an abrupt discontinuation of chronic therapies including benzodiazepines. The withdrawal symptomatology may also include insomnia, irritability, muscular pain and stiffness, hand tremor, anxiety, and psychotic episodes [1]. An acute reduction in brain gamma-aminobutyric acid function has been indicated as a pathogenic mechanism of the syndrome [2]. While generalized tonic–clonic seizures are the most common, focal seizures are rarely reported in this condition. Particularly, focal nonconvulsive seizures have been described in patients treated with flumazenil during detoxification from lormetazepam abuse [3].
2 Case report
A 46-year-old woman with a clinical history of schizoaffective and personality disorders presented to our hospital with new-onset convulsive seizures. It was reported that the patient was forced to interrupt chronic lormetazepam abuse (a daily drug consumption of 150–250 mg of an oral solution) the day before the onset of seizures. When she ran out of the medication, she ingested an oral solution containing 800 mg of amitriptyline as a supposed sedative substitute for lormetazepam. Convulsive seizures recurred multiple times per day in the hospital during daytime and night-time hours in a clinical context of insomnia, restlessness, muscular rigidity, and painful spasms affecting the back muscles that implicated a benzodiazepine withdrawal syndrome. There was no evidence of amitriptyline toxicity, as the patient did not show sedation, anticholinergic effects, arterial hypotension, or cardiac arrhythmias. The amitriptyline blood level was not obtained. Seizures presented with a duration of approximately 53 s and a paroxysmal, stereotyped sequence of movements that was documented with video-EEG. The recordings of seizures, using the usual 10–20 system of electrode placement, did not show EEG abnormalities or postictal slowing of the background activity, but the tracings were disturbed by myogenic artifacts throughout the course of the events (Fig. 1, Fig. 2). Awareness was preserved during the episodes. The patient occasionally called for help at seizure onset but remained unresponsive until the cessation of movements. Rapid recovery of speech and an absence of postictal confusion were reported on different occasions. Magnetic resonance imaging of the brain was normal. Three days after the patient's admission, a cluster of convulsive episodes prompted a short intensive treatment with propofol and mechanical ventilation. Seizures were increasingly observed again after extubation, despite oral treatment with clonazepam, levetiracetam, and repeated intravenous administrations of diazepam. Regardless of the inefficacy of the treatment during an eight-day period of intense seizure activity, the patient eventually became seizure-free. The remission of seizures occurred simultaneously with the cessation of insomnia and muscular symptoms related to lormetazepam withdrawal. Clonazepam and levetiracetam were subsequently reduced and discontinued as the patient remained asymptomatic. Despite the absence of ictal EEG abnormalities, a retrospective analysis of the seizure semiology based on video recordings and additional recorded observations of the episodes at bedside suggested the clinical diagnosis of FBMS, presumably arising from the supplementary sensorimotor area (SSMA). This diagnostic conclusion was supported by the evidence of a paroxysmal, stereotyped ictal motor pattern with early abduction of the arm flexed at the elbow and subsequent asymmetric tonic posturing of the upper limb that reproduced the figure four sign [4] at seizure onset. The left arm was extended, and the right arm was flexed (at the elbow), while the head was briefly turned to the left at this point of the seizure. Consequent bilateral clonic movements of the upper limbs showed a typical epileptic pattern with a gradual frequency decline and a progressive amplitude increase of movements [5]. A stereotyped left hemifacial spasm was also noted during the tonic contraction of the upper limbs. The short duration of the episodes, occasional evidence of cyanosis and lack of pupillary reactivity served as additional clues to the ES diagnosis. Although the clinical picture was initially confounded by the patient's preserved awareness during the bilateral convulsions, the absence of postictal confusion, and the normal EEG findings, these features finally appeared consistent with FBMS arising in the SSMA, as previously reported in the literature [[6], [7], [8]]. The patient was discharged with a normal neurologic status nine days after the last convulsive episode. Seizures did not recur even without antiseizure medication during the following month, after which the patient was lost to follow-up.Fig. 1 EEG recording of the initial phase of a focal bilateral motor seizure with a still-frame from the video.
EEG recording of the initial phase of a focal bilateral motor seizure with a still-frame from the video.
The EEG does not reveal evident abnormalities at seizure onset. The tracing is progressively disturbed by myogenic artifacts during the evolution of the seizure motor pattern.
Fig. 1Fig. 2 EEG recording of the final phase of a focal bilateral motor seizure.
Phasic myogenic artifacts mark the tracing during the ending clonic phase of the seizure. The EEG shows the persistence of the background activity in the immediate postictal period.
Fig. 2
3 Discussion
The manifestation of convulsive seizures without EEG abnormalities in the context of benzodiazepine withdrawal syndrome and psychiatric disease requires a differential diagnosis of ES and nonepileptic seizures (NES). In fact, the abrupt interruption of high dosage, chronic benzodiazepine treatment represents a potential precipitating factor for ES, while NES are occasionally observed as imitators of epilepsy in patients with mental health problems [9,10]. In addition to the psychiatric disorders and the normal neurophysiologic findings, the patient's preserved awareness during the bilateral convulsions and the absence of postictal confusion confounded the clinical picture with the potential to misdiagnose FBMS as NES. FBMS are described as motor onset seizures arising in one cerebral hemisphere and rapidly involving bilateral motor networks, often with asymmetric tonic posturing [11]. The SSMA is classically recognized as the anatomical origin of FBMS [11]. Clinical characteristics of focal onset seizures arising in the SSMA include short duration (typically not exceeding 40 s) [6,12]; predominant onset during sleep [6,8]; somatosensory aura [8,13]; preserved awareness [6,8,13]; absence of postictal confusion [8]; vocalization or speech arrest [6,8,13]; unilateral tonic posturing [8,13] or bilateral, asymmetric tonic posturing [4,13,14]; bilateral abduction of the upper extremities [6]; and bilateral thrashing movements [6,8] that usually constitute a paroxysmal, stereotyped motor pattern in the same individual [6]. Particularly, asymmetric tonic posturing has been related to the asynchronous activation of the SSMA [4,14], thus representing a key semiologic feature of seizures involving this cortical area [8,12]. Authors have also referred to asymmetric tonic posturing as the figure four sign that marks focal to bilateral tonic–clonic seizures [4]. The lateralizing value of this sign has been repeatedly reported with the epileptogenic focus usually contralateral to the arm that is extended at the elbow [4,14]. In contrast, NES with convulsive-like manifestations are characterized by long duration (frequently exceeding 2 min); fluctuating course; and asynchronous movements (typically side-to-side head movements and out of phase limb movements) that show unchanged frequency and variable amplitude throughout the event [10].
In our patient, the convulsive disorders shared most of the mentioned clinical characteristics of focal onset seizures arising in the SSMA. In particular, the origin of ictal symptoms in the SSMA was suggested by the early stereotyped contraction of the left arm and consequent asymmetric tonic posturing of limbs, while a left hemifacial spasm appeared consistent with the presumed epileptic focus in the right cerebral hemisphere. The progression of these phenomena to bilateral, tonic extension and clonic movements of the upper extremities characterized the final phase of the motor pattern mimicking focal to bilateral tonic–clonic seizures. However, the absence of awareness impairments, postictal confusion and EEG slowing in the immediate postictal phase discredited the apparent bilateral tonic–clonic evolution of focal seizures, suggesting limited epileptic involvement of the SSMA during the entire course of the events. While the seizure semiology could be referred to FBMS arising from the SSMA, the absence of ictal EEG abnormalities to confirm epileptic seizures is a limiting factor for the diagnosis of frontal lobe seizures. This suggests the potential utility of EEG recordings with additional midline electrodes (placed according to the 10–10 system) and/or ictal SPECT, as a supplementary method to confirm the epileptic nature and the anatomical origin of the seizures. However, it is not uncommon for seizures arising in the mesial and deep structures of the frontal lobe to have absent scalp ictal EEG changes utilizing the 10-20 international system of electrode placement.
Although structural brain abnormalities were not documented in our patient, neurochemical alterations induced by benzodiazepine withdrawal into specific networks may have resulted in a major seizure susceptibility of the SSMA. Interestingly, a major intrinsic epileptogenicity of specific cortical areas has been postulated by Albiero et al. to explain the occurrence of focal seizures in patients who underwent detoxification from chronic lormetazepam abuse [3]. Additional information is needed to ascertain the major occurrence of focal seizures in patients who interrupted chronic lormetazepam abuse. Although amitriptyline may have lowered the seizure threshold, the precise clinical impact of the substitution remains uncertain in our patient, as typical symptoms of amitriptyline intoxication such as sedation, arterial hypotension, cardiac arrhythmias or anticholinergic effects [15] were not observed during the period of recurrent seizures. In contrast, the strict concomitance of seizures with classic benzodiazepine abstinence symptoms such as insomnia, restlessness, rigidity, and muscular pain clearly indicated the primary precipitating role of lormetazepam withdrawal. A previous report remarked that lormetazepam is the benzodiazepine that is most frequently associated with abuse and dependence [16], particularly when it is used in the form of an oral solution [17]. The presence of ethanol in the oral solution of lormetazepam has been specifically suspected to increase the risks for abuse and dependence. Although ethanol quantities may not be relevant in patients treated with accurate doses of lormetazepam (1.5–2.5 mg daily for insomnia), the pathogenic effect of ethanol may be significant in lormetazepam addicted patients. The present case provides additional information about the risks of lormetazepam addiction as a component of psychiatric care. We suggest strict monitoring of patients treated with an oral solution of this medication.
4 Conclusion
FBMS may be observed during the clinical manifestation of lormetazepam withdrawal in patients who overdosed on amitriptyline. FBMS may produce a confusing clinical picture, with the potential to misdiagnose NES, especially when no abnormal neurophysiologic findings are clearly identified and when patients have a known underlying psychiatric disease.
The following is the supplementary data related to this article.Video 1
Focal bilateral motor seizure precipitated by lormetazepam withdrawal and amitriptyline overdose.
The seizure begins with a tonic posture of the left hand and abduction of the left arm flexed at the elbow. The left arm is then elevated and extended anteriorly, while the right arm is initially flexed at the elbow and rotated inward, reproducing the figure four sign. The head is briefly turned to the left during the asymmetric posturing of the upper extremities. Consequently, the right arm is also extended anteriorly. A slight facial cyanosis and a grimace with left hemifacial spasm (consistent with the presumed origin of the seizure in the right cerebral hemisphere) are noted during the extension of the upper limbs. There is no eye blink at the threatening gesture or pupillary response to the light at this point of the seizure. Convulsions cease with a sequence of vibratory and clonic movements of the upper extremities and trunk. The duration of the seizure is approximately 53 s and amitriptyline overdose.
Video 1
Ethical statement
The authors state that the study described in the paper complies with the publishing ethics of the journal.
The patient’s face has been partially obscured in the attached video, as requested by the patient herself at the time of the given consent.
Declaration of competing interest
The authors state that there is no conflict of interest. | Oral | DrugAdministrationRoute | CC BY-NC-ND | 33490945 | 18,821,337 | 2021 |
What was the administration route of drug 'LORMETAZEPAM'? | Focal bilateral motor seizures precipitated by abrupt cessation of chronic lormetazepam abuse and amitriptyline overdose.
We report the case of an adult psychiatric patient who developed new-onset focal bilateral motor seizures (FBMS) in the context of a severe benzodiazepine withdrawal syndrome. The patient was forced to interrupt chronic lormetazepam abuse and overdosed on amitriptyline (800 mg in an oral solution) before the onset of seizures. Typical signs of amitriptyline intoxication such as sedation and anticholinergic effects were not observed. Video-EEG recordings revealed a stereotypical ictal motor pattern with asymmetric tonic posturing and bilateral clonic movements of the upper limbs, but there were no abnormalities identified by EEG. Seizures recurred multiple times per day but resolved simultaneously when withdrawal symptomatology subsided eight days after onset. Nonepileptic seizures (NES) were considered in the differential diagnosis because of the patient's psychiatric history including preserved awareness during the bilateral convulsions, the absence of postictal confusion, and normal EEG. The present case indicates that FBMS may occur during benzodiazepine withdrawal in patients who overdosed on amitriptyline. The diagnosis may be challenging as FBMS may mimic NES in the absence of abnormal neurophysiologic findings. This may be especially challenging in patients with an underlying psychiatric disease.
1 Introduction
Epileptic seizures (ES) have been frequently reported as clinical manifestations of benzodiazepine withdrawal, an acute disorder precipitated by an abrupt discontinuation of chronic therapies including benzodiazepines. The withdrawal symptomatology may also include insomnia, irritability, muscular pain and stiffness, hand tremor, anxiety, and psychotic episodes [1]. An acute reduction in brain gamma-aminobutyric acid function has been indicated as a pathogenic mechanism of the syndrome [2]. While generalized tonic–clonic seizures are the most common, focal seizures are rarely reported in this condition. Particularly, focal nonconvulsive seizures have been described in patients treated with flumazenil during detoxification from lormetazepam abuse [3].
2 Case report
A 46-year-old woman with a clinical history of schizoaffective and personality disorders presented to our hospital with new-onset convulsive seizures. It was reported that the patient was forced to interrupt chronic lormetazepam abuse (a daily drug consumption of 150–250 mg of an oral solution) the day before the onset of seizures. When she ran out of the medication, she ingested an oral solution containing 800 mg of amitriptyline as a supposed sedative substitute for lormetazepam. Convulsive seizures recurred multiple times per day in the hospital during daytime and night-time hours in a clinical context of insomnia, restlessness, muscular rigidity, and painful spasms affecting the back muscles that implicated a benzodiazepine withdrawal syndrome. There was no evidence of amitriptyline toxicity, as the patient did not show sedation, anticholinergic effects, arterial hypotension, or cardiac arrhythmias. The amitriptyline blood level was not obtained. Seizures presented with a duration of approximately 53 s and a paroxysmal, stereotyped sequence of movements that was documented with video-EEG. The recordings of seizures, using the usual 10–20 system of electrode placement, did not show EEG abnormalities or postictal slowing of the background activity, but the tracings were disturbed by myogenic artifacts throughout the course of the events (Fig. 1, Fig. 2). Awareness was preserved during the episodes. The patient occasionally called for help at seizure onset but remained unresponsive until the cessation of movements. Rapid recovery of speech and an absence of postictal confusion were reported on different occasions. Magnetic resonance imaging of the brain was normal. Three days after the patient's admission, a cluster of convulsive episodes prompted a short intensive treatment with propofol and mechanical ventilation. Seizures were increasingly observed again after extubation, despite oral treatment with clonazepam, levetiracetam, and repeated intravenous administrations of diazepam. Regardless of the inefficacy of the treatment during an eight-day period of intense seizure activity, the patient eventually became seizure-free. The remission of seizures occurred simultaneously with the cessation of insomnia and muscular symptoms related to lormetazepam withdrawal. Clonazepam and levetiracetam were subsequently reduced and discontinued as the patient remained asymptomatic. Despite the absence of ictal EEG abnormalities, a retrospective analysis of the seizure semiology based on video recordings and additional recorded observations of the episodes at bedside suggested the clinical diagnosis of FBMS, presumably arising from the supplementary sensorimotor area (SSMA). This diagnostic conclusion was supported by the evidence of a paroxysmal, stereotyped ictal motor pattern with early abduction of the arm flexed at the elbow and subsequent asymmetric tonic posturing of the upper limb that reproduced the figure four sign [4] at seizure onset. The left arm was extended, and the right arm was flexed (at the elbow), while the head was briefly turned to the left at this point of the seizure. Consequent bilateral clonic movements of the upper limbs showed a typical epileptic pattern with a gradual frequency decline and a progressive amplitude increase of movements [5]. A stereotyped left hemifacial spasm was also noted during the tonic contraction of the upper limbs. The short duration of the episodes, occasional evidence of cyanosis and lack of pupillary reactivity served as additional clues to the ES diagnosis. Although the clinical picture was initially confounded by the patient's preserved awareness during the bilateral convulsions, the absence of postictal confusion, and the normal EEG findings, these features finally appeared consistent with FBMS arising in the SSMA, as previously reported in the literature [[6], [7], [8]]. The patient was discharged with a normal neurologic status nine days after the last convulsive episode. Seizures did not recur even without antiseizure medication during the following month, after which the patient was lost to follow-up.Fig. 1 EEG recording of the initial phase of a focal bilateral motor seizure with a still-frame from the video.
EEG recording of the initial phase of a focal bilateral motor seizure with a still-frame from the video.
The EEG does not reveal evident abnormalities at seizure onset. The tracing is progressively disturbed by myogenic artifacts during the evolution of the seizure motor pattern.
Fig. 1Fig. 2 EEG recording of the final phase of a focal bilateral motor seizure.
Phasic myogenic artifacts mark the tracing during the ending clonic phase of the seizure. The EEG shows the persistence of the background activity in the immediate postictal period.
Fig. 2
3 Discussion
The manifestation of convulsive seizures without EEG abnormalities in the context of benzodiazepine withdrawal syndrome and psychiatric disease requires a differential diagnosis of ES and nonepileptic seizures (NES). In fact, the abrupt interruption of high dosage, chronic benzodiazepine treatment represents a potential precipitating factor for ES, while NES are occasionally observed as imitators of epilepsy in patients with mental health problems [9,10]. In addition to the psychiatric disorders and the normal neurophysiologic findings, the patient's preserved awareness during the bilateral convulsions and the absence of postictal confusion confounded the clinical picture with the potential to misdiagnose FBMS as NES. FBMS are described as motor onset seizures arising in one cerebral hemisphere and rapidly involving bilateral motor networks, often with asymmetric tonic posturing [11]. The SSMA is classically recognized as the anatomical origin of FBMS [11]. Clinical characteristics of focal onset seizures arising in the SSMA include short duration (typically not exceeding 40 s) [6,12]; predominant onset during sleep [6,8]; somatosensory aura [8,13]; preserved awareness [6,8,13]; absence of postictal confusion [8]; vocalization or speech arrest [6,8,13]; unilateral tonic posturing [8,13] or bilateral, asymmetric tonic posturing [4,13,14]; bilateral abduction of the upper extremities [6]; and bilateral thrashing movements [6,8] that usually constitute a paroxysmal, stereotyped motor pattern in the same individual [6]. Particularly, asymmetric tonic posturing has been related to the asynchronous activation of the SSMA [4,14], thus representing a key semiologic feature of seizures involving this cortical area [8,12]. Authors have also referred to asymmetric tonic posturing as the figure four sign that marks focal to bilateral tonic–clonic seizures [4]. The lateralizing value of this sign has been repeatedly reported with the epileptogenic focus usually contralateral to the arm that is extended at the elbow [4,14]. In contrast, NES with convulsive-like manifestations are characterized by long duration (frequently exceeding 2 min); fluctuating course; and asynchronous movements (typically side-to-side head movements and out of phase limb movements) that show unchanged frequency and variable amplitude throughout the event [10].
In our patient, the convulsive disorders shared most of the mentioned clinical characteristics of focal onset seizures arising in the SSMA. In particular, the origin of ictal symptoms in the SSMA was suggested by the early stereotyped contraction of the left arm and consequent asymmetric tonic posturing of limbs, while a left hemifacial spasm appeared consistent with the presumed epileptic focus in the right cerebral hemisphere. The progression of these phenomena to bilateral, tonic extension and clonic movements of the upper extremities characterized the final phase of the motor pattern mimicking focal to bilateral tonic–clonic seizures. However, the absence of awareness impairments, postictal confusion and EEG slowing in the immediate postictal phase discredited the apparent bilateral tonic–clonic evolution of focal seizures, suggesting limited epileptic involvement of the SSMA during the entire course of the events. While the seizure semiology could be referred to FBMS arising from the SSMA, the absence of ictal EEG abnormalities to confirm epileptic seizures is a limiting factor for the diagnosis of frontal lobe seizures. This suggests the potential utility of EEG recordings with additional midline electrodes (placed according to the 10–10 system) and/or ictal SPECT, as a supplementary method to confirm the epileptic nature and the anatomical origin of the seizures. However, it is not uncommon for seizures arising in the mesial and deep structures of the frontal lobe to have absent scalp ictal EEG changes utilizing the 10-20 international system of electrode placement.
Although structural brain abnormalities were not documented in our patient, neurochemical alterations induced by benzodiazepine withdrawal into specific networks may have resulted in a major seizure susceptibility of the SSMA. Interestingly, a major intrinsic epileptogenicity of specific cortical areas has been postulated by Albiero et al. to explain the occurrence of focal seizures in patients who underwent detoxification from chronic lormetazepam abuse [3]. Additional information is needed to ascertain the major occurrence of focal seizures in patients who interrupted chronic lormetazepam abuse. Although amitriptyline may have lowered the seizure threshold, the precise clinical impact of the substitution remains uncertain in our patient, as typical symptoms of amitriptyline intoxication such as sedation, arterial hypotension, cardiac arrhythmias or anticholinergic effects [15] were not observed during the period of recurrent seizures. In contrast, the strict concomitance of seizures with classic benzodiazepine abstinence symptoms such as insomnia, restlessness, rigidity, and muscular pain clearly indicated the primary precipitating role of lormetazepam withdrawal. A previous report remarked that lormetazepam is the benzodiazepine that is most frequently associated with abuse and dependence [16], particularly when it is used in the form of an oral solution [17]. The presence of ethanol in the oral solution of lormetazepam has been specifically suspected to increase the risks for abuse and dependence. Although ethanol quantities may not be relevant in patients treated with accurate doses of lormetazepam (1.5–2.5 mg daily for insomnia), the pathogenic effect of ethanol may be significant in lormetazepam addicted patients. The present case provides additional information about the risks of lormetazepam addiction as a component of psychiatric care. We suggest strict monitoring of patients treated with an oral solution of this medication.
4 Conclusion
FBMS may be observed during the clinical manifestation of lormetazepam withdrawal in patients who overdosed on amitriptyline. FBMS may produce a confusing clinical picture, with the potential to misdiagnose NES, especially when no abnormal neurophysiologic findings are clearly identified and when patients have a known underlying psychiatric disease.
The following is the supplementary data related to this article.Video 1
Focal bilateral motor seizure precipitated by lormetazepam withdrawal and amitriptyline overdose.
The seizure begins with a tonic posture of the left hand and abduction of the left arm flexed at the elbow. The left arm is then elevated and extended anteriorly, while the right arm is initially flexed at the elbow and rotated inward, reproducing the figure four sign. The head is briefly turned to the left during the asymmetric posturing of the upper extremities. Consequently, the right arm is also extended anteriorly. A slight facial cyanosis and a grimace with left hemifacial spasm (consistent with the presumed origin of the seizure in the right cerebral hemisphere) are noted during the extension of the upper limbs. There is no eye blink at the threatening gesture or pupillary response to the light at this point of the seizure. Convulsions cease with a sequence of vibratory and clonic movements of the upper extremities and trunk. The duration of the seizure is approximately 53 s and amitriptyline overdose.
Video 1
Ethical statement
The authors state that the study described in the paper complies with the publishing ethics of the journal.
The patient’s face has been partially obscured in the attached video, as requested by the patient herself at the time of the given consent.
Declaration of competing interest
The authors state that there is no conflict of interest. | Oral | DrugAdministrationRoute | CC BY-NC-ND | 33490945 | 18,821,337 | 2021 |
What was the dosage of drug 'DIAZEPAM'? | Focal bilateral motor seizures precipitated by abrupt cessation of chronic lormetazepam abuse and amitriptyline overdose.
We report the case of an adult psychiatric patient who developed new-onset focal bilateral motor seizures (FBMS) in the context of a severe benzodiazepine withdrawal syndrome. The patient was forced to interrupt chronic lormetazepam abuse and overdosed on amitriptyline (800 mg in an oral solution) before the onset of seizures. Typical signs of amitriptyline intoxication such as sedation and anticholinergic effects were not observed. Video-EEG recordings revealed a stereotypical ictal motor pattern with asymmetric tonic posturing and bilateral clonic movements of the upper limbs, but there were no abnormalities identified by EEG. Seizures recurred multiple times per day but resolved simultaneously when withdrawal symptomatology subsided eight days after onset. Nonepileptic seizures (NES) were considered in the differential diagnosis because of the patient's psychiatric history including preserved awareness during the bilateral convulsions, the absence of postictal confusion, and normal EEG. The present case indicates that FBMS may occur during benzodiazepine withdrawal in patients who overdosed on amitriptyline. The diagnosis may be challenging as FBMS may mimic NES in the absence of abnormal neurophysiologic findings. This may be especially challenging in patients with an underlying psychiatric disease.
1 Introduction
Epileptic seizures (ES) have been frequently reported as clinical manifestations of benzodiazepine withdrawal, an acute disorder precipitated by an abrupt discontinuation of chronic therapies including benzodiazepines. The withdrawal symptomatology may also include insomnia, irritability, muscular pain and stiffness, hand tremor, anxiety, and psychotic episodes [1]. An acute reduction in brain gamma-aminobutyric acid function has been indicated as a pathogenic mechanism of the syndrome [2]. While generalized tonic–clonic seizures are the most common, focal seizures are rarely reported in this condition. Particularly, focal nonconvulsive seizures have been described in patients treated with flumazenil during detoxification from lormetazepam abuse [3].
2 Case report
A 46-year-old woman with a clinical history of schizoaffective and personality disorders presented to our hospital with new-onset convulsive seizures. It was reported that the patient was forced to interrupt chronic lormetazepam abuse (a daily drug consumption of 150–250 mg of an oral solution) the day before the onset of seizures. When she ran out of the medication, she ingested an oral solution containing 800 mg of amitriptyline as a supposed sedative substitute for lormetazepam. Convulsive seizures recurred multiple times per day in the hospital during daytime and night-time hours in a clinical context of insomnia, restlessness, muscular rigidity, and painful spasms affecting the back muscles that implicated a benzodiazepine withdrawal syndrome. There was no evidence of amitriptyline toxicity, as the patient did not show sedation, anticholinergic effects, arterial hypotension, or cardiac arrhythmias. The amitriptyline blood level was not obtained. Seizures presented with a duration of approximately 53 s and a paroxysmal, stereotyped sequence of movements that was documented with video-EEG. The recordings of seizures, using the usual 10–20 system of electrode placement, did not show EEG abnormalities or postictal slowing of the background activity, but the tracings were disturbed by myogenic artifacts throughout the course of the events (Fig. 1, Fig. 2). Awareness was preserved during the episodes. The patient occasionally called for help at seizure onset but remained unresponsive until the cessation of movements. Rapid recovery of speech and an absence of postictal confusion were reported on different occasions. Magnetic resonance imaging of the brain was normal. Three days after the patient's admission, a cluster of convulsive episodes prompted a short intensive treatment with propofol and mechanical ventilation. Seizures were increasingly observed again after extubation, despite oral treatment with clonazepam, levetiracetam, and repeated intravenous administrations of diazepam. Regardless of the inefficacy of the treatment during an eight-day period of intense seizure activity, the patient eventually became seizure-free. The remission of seizures occurred simultaneously with the cessation of insomnia and muscular symptoms related to lormetazepam withdrawal. Clonazepam and levetiracetam were subsequently reduced and discontinued as the patient remained asymptomatic. Despite the absence of ictal EEG abnormalities, a retrospective analysis of the seizure semiology based on video recordings and additional recorded observations of the episodes at bedside suggested the clinical diagnosis of FBMS, presumably arising from the supplementary sensorimotor area (SSMA). This diagnostic conclusion was supported by the evidence of a paroxysmal, stereotyped ictal motor pattern with early abduction of the arm flexed at the elbow and subsequent asymmetric tonic posturing of the upper limb that reproduced the figure four sign [4] at seizure onset. The left arm was extended, and the right arm was flexed (at the elbow), while the head was briefly turned to the left at this point of the seizure. Consequent bilateral clonic movements of the upper limbs showed a typical epileptic pattern with a gradual frequency decline and a progressive amplitude increase of movements [5]. A stereotyped left hemifacial spasm was also noted during the tonic contraction of the upper limbs. The short duration of the episodes, occasional evidence of cyanosis and lack of pupillary reactivity served as additional clues to the ES diagnosis. Although the clinical picture was initially confounded by the patient's preserved awareness during the bilateral convulsions, the absence of postictal confusion, and the normal EEG findings, these features finally appeared consistent with FBMS arising in the SSMA, as previously reported in the literature [[6], [7], [8]]. The patient was discharged with a normal neurologic status nine days after the last convulsive episode. Seizures did not recur even without antiseizure medication during the following month, after which the patient was lost to follow-up.Fig. 1 EEG recording of the initial phase of a focal bilateral motor seizure with a still-frame from the video.
EEG recording of the initial phase of a focal bilateral motor seizure with a still-frame from the video.
The EEG does not reveal evident abnormalities at seizure onset. The tracing is progressively disturbed by myogenic artifacts during the evolution of the seizure motor pattern.
Fig. 1Fig. 2 EEG recording of the final phase of a focal bilateral motor seizure.
Phasic myogenic artifacts mark the tracing during the ending clonic phase of the seizure. The EEG shows the persistence of the background activity in the immediate postictal period.
Fig. 2
3 Discussion
The manifestation of convulsive seizures without EEG abnormalities in the context of benzodiazepine withdrawal syndrome and psychiatric disease requires a differential diagnosis of ES and nonepileptic seizures (NES). In fact, the abrupt interruption of high dosage, chronic benzodiazepine treatment represents a potential precipitating factor for ES, while NES are occasionally observed as imitators of epilepsy in patients with mental health problems [9,10]. In addition to the psychiatric disorders and the normal neurophysiologic findings, the patient's preserved awareness during the bilateral convulsions and the absence of postictal confusion confounded the clinical picture with the potential to misdiagnose FBMS as NES. FBMS are described as motor onset seizures arising in one cerebral hemisphere and rapidly involving bilateral motor networks, often with asymmetric tonic posturing [11]. The SSMA is classically recognized as the anatomical origin of FBMS [11]. Clinical characteristics of focal onset seizures arising in the SSMA include short duration (typically not exceeding 40 s) [6,12]; predominant onset during sleep [6,8]; somatosensory aura [8,13]; preserved awareness [6,8,13]; absence of postictal confusion [8]; vocalization or speech arrest [6,8,13]; unilateral tonic posturing [8,13] or bilateral, asymmetric tonic posturing [4,13,14]; bilateral abduction of the upper extremities [6]; and bilateral thrashing movements [6,8] that usually constitute a paroxysmal, stereotyped motor pattern in the same individual [6]. Particularly, asymmetric tonic posturing has been related to the asynchronous activation of the SSMA [4,14], thus representing a key semiologic feature of seizures involving this cortical area [8,12]. Authors have also referred to asymmetric tonic posturing as the figure four sign that marks focal to bilateral tonic–clonic seizures [4]. The lateralizing value of this sign has been repeatedly reported with the epileptogenic focus usually contralateral to the arm that is extended at the elbow [4,14]. In contrast, NES with convulsive-like manifestations are characterized by long duration (frequently exceeding 2 min); fluctuating course; and asynchronous movements (typically side-to-side head movements and out of phase limb movements) that show unchanged frequency and variable amplitude throughout the event [10].
In our patient, the convulsive disorders shared most of the mentioned clinical characteristics of focal onset seizures arising in the SSMA. In particular, the origin of ictal symptoms in the SSMA was suggested by the early stereotyped contraction of the left arm and consequent asymmetric tonic posturing of limbs, while a left hemifacial spasm appeared consistent with the presumed epileptic focus in the right cerebral hemisphere. The progression of these phenomena to bilateral, tonic extension and clonic movements of the upper extremities characterized the final phase of the motor pattern mimicking focal to bilateral tonic–clonic seizures. However, the absence of awareness impairments, postictal confusion and EEG slowing in the immediate postictal phase discredited the apparent bilateral tonic–clonic evolution of focal seizures, suggesting limited epileptic involvement of the SSMA during the entire course of the events. While the seizure semiology could be referred to FBMS arising from the SSMA, the absence of ictal EEG abnormalities to confirm epileptic seizures is a limiting factor for the diagnosis of frontal lobe seizures. This suggests the potential utility of EEG recordings with additional midline electrodes (placed according to the 10–10 system) and/or ictal SPECT, as a supplementary method to confirm the epileptic nature and the anatomical origin of the seizures. However, it is not uncommon for seizures arising in the mesial and deep structures of the frontal lobe to have absent scalp ictal EEG changes utilizing the 10-20 international system of electrode placement.
Although structural brain abnormalities were not documented in our patient, neurochemical alterations induced by benzodiazepine withdrawal into specific networks may have resulted in a major seizure susceptibility of the SSMA. Interestingly, a major intrinsic epileptogenicity of specific cortical areas has been postulated by Albiero et al. to explain the occurrence of focal seizures in patients who underwent detoxification from chronic lormetazepam abuse [3]. Additional information is needed to ascertain the major occurrence of focal seizures in patients who interrupted chronic lormetazepam abuse. Although amitriptyline may have lowered the seizure threshold, the precise clinical impact of the substitution remains uncertain in our patient, as typical symptoms of amitriptyline intoxication such as sedation, arterial hypotension, cardiac arrhythmias or anticholinergic effects [15] were not observed during the period of recurrent seizures. In contrast, the strict concomitance of seizures with classic benzodiazepine abstinence symptoms such as insomnia, restlessness, rigidity, and muscular pain clearly indicated the primary precipitating role of lormetazepam withdrawal. A previous report remarked that lormetazepam is the benzodiazepine that is most frequently associated with abuse and dependence [16], particularly when it is used in the form of an oral solution [17]. The presence of ethanol in the oral solution of lormetazepam has been specifically suspected to increase the risks for abuse and dependence. Although ethanol quantities may not be relevant in patients treated with accurate doses of lormetazepam (1.5–2.5 mg daily for insomnia), the pathogenic effect of ethanol may be significant in lormetazepam addicted patients. The present case provides additional information about the risks of lormetazepam addiction as a component of psychiatric care. We suggest strict monitoring of patients treated with an oral solution of this medication.
4 Conclusion
FBMS may be observed during the clinical manifestation of lormetazepam withdrawal in patients who overdosed on amitriptyline. FBMS may produce a confusing clinical picture, with the potential to misdiagnose NES, especially when no abnormal neurophysiologic findings are clearly identified and when patients have a known underlying psychiatric disease.
The following is the supplementary data related to this article.Video 1
Focal bilateral motor seizure precipitated by lormetazepam withdrawal and amitriptyline overdose.
The seizure begins with a tonic posture of the left hand and abduction of the left arm flexed at the elbow. The left arm is then elevated and extended anteriorly, while the right arm is initially flexed at the elbow and rotated inward, reproducing the figure four sign. The head is briefly turned to the left during the asymmetric posturing of the upper extremities. Consequently, the right arm is also extended anteriorly. A slight facial cyanosis and a grimace with left hemifacial spasm (consistent with the presumed origin of the seizure in the right cerebral hemisphere) are noted during the extension of the upper limbs. There is no eye blink at the threatening gesture or pupillary response to the light at this point of the seizure. Convulsions cease with a sequence of vibratory and clonic movements of the upper extremities and trunk. The duration of the seizure is approximately 53 s and amitriptyline overdose.
Video 1
Ethical statement
The authors state that the study described in the paper complies with the publishing ethics of the journal.
The patient’s face has been partially obscured in the attached video, as requested by the patient herself at the time of the given consent.
Declaration of competing interest
The authors state that there is no conflict of interest. | REPEATED IV TREATMENT | DrugDosageText | CC BY-NC-ND | 33490945 | 18,821,337 | 2021 |
What was the outcome of reaction 'Drug abuse'? | Focal bilateral motor seizures precipitated by abrupt cessation of chronic lormetazepam abuse and amitriptyline overdose.
We report the case of an adult psychiatric patient who developed new-onset focal bilateral motor seizures (FBMS) in the context of a severe benzodiazepine withdrawal syndrome. The patient was forced to interrupt chronic lormetazepam abuse and overdosed on amitriptyline (800 mg in an oral solution) before the onset of seizures. Typical signs of amitriptyline intoxication such as sedation and anticholinergic effects were not observed. Video-EEG recordings revealed a stereotypical ictal motor pattern with asymmetric tonic posturing and bilateral clonic movements of the upper limbs, but there were no abnormalities identified by EEG. Seizures recurred multiple times per day but resolved simultaneously when withdrawal symptomatology subsided eight days after onset. Nonepileptic seizures (NES) were considered in the differential diagnosis because of the patient's psychiatric history including preserved awareness during the bilateral convulsions, the absence of postictal confusion, and normal EEG. The present case indicates that FBMS may occur during benzodiazepine withdrawal in patients who overdosed on amitriptyline. The diagnosis may be challenging as FBMS may mimic NES in the absence of abnormal neurophysiologic findings. This may be especially challenging in patients with an underlying psychiatric disease.
1 Introduction
Epileptic seizures (ES) have been frequently reported as clinical manifestations of benzodiazepine withdrawal, an acute disorder precipitated by an abrupt discontinuation of chronic therapies including benzodiazepines. The withdrawal symptomatology may also include insomnia, irritability, muscular pain and stiffness, hand tremor, anxiety, and psychotic episodes [1]. An acute reduction in brain gamma-aminobutyric acid function has been indicated as a pathogenic mechanism of the syndrome [2]. While generalized tonic–clonic seizures are the most common, focal seizures are rarely reported in this condition. Particularly, focal nonconvulsive seizures have been described in patients treated with flumazenil during detoxification from lormetazepam abuse [3].
2 Case report
A 46-year-old woman with a clinical history of schizoaffective and personality disorders presented to our hospital with new-onset convulsive seizures. It was reported that the patient was forced to interrupt chronic lormetazepam abuse (a daily drug consumption of 150–250 mg of an oral solution) the day before the onset of seizures. When she ran out of the medication, she ingested an oral solution containing 800 mg of amitriptyline as a supposed sedative substitute for lormetazepam. Convulsive seizures recurred multiple times per day in the hospital during daytime and night-time hours in a clinical context of insomnia, restlessness, muscular rigidity, and painful spasms affecting the back muscles that implicated a benzodiazepine withdrawal syndrome. There was no evidence of amitriptyline toxicity, as the patient did not show sedation, anticholinergic effects, arterial hypotension, or cardiac arrhythmias. The amitriptyline blood level was not obtained. Seizures presented with a duration of approximately 53 s and a paroxysmal, stereotyped sequence of movements that was documented with video-EEG. The recordings of seizures, using the usual 10–20 system of electrode placement, did not show EEG abnormalities or postictal slowing of the background activity, but the tracings were disturbed by myogenic artifacts throughout the course of the events (Fig. 1, Fig. 2). Awareness was preserved during the episodes. The patient occasionally called for help at seizure onset but remained unresponsive until the cessation of movements. Rapid recovery of speech and an absence of postictal confusion were reported on different occasions. Magnetic resonance imaging of the brain was normal. Three days after the patient's admission, a cluster of convulsive episodes prompted a short intensive treatment with propofol and mechanical ventilation. Seizures were increasingly observed again after extubation, despite oral treatment with clonazepam, levetiracetam, and repeated intravenous administrations of diazepam. Regardless of the inefficacy of the treatment during an eight-day period of intense seizure activity, the patient eventually became seizure-free. The remission of seizures occurred simultaneously with the cessation of insomnia and muscular symptoms related to lormetazepam withdrawal. Clonazepam and levetiracetam were subsequently reduced and discontinued as the patient remained asymptomatic. Despite the absence of ictal EEG abnormalities, a retrospective analysis of the seizure semiology based on video recordings and additional recorded observations of the episodes at bedside suggested the clinical diagnosis of FBMS, presumably arising from the supplementary sensorimotor area (SSMA). This diagnostic conclusion was supported by the evidence of a paroxysmal, stereotyped ictal motor pattern with early abduction of the arm flexed at the elbow and subsequent asymmetric tonic posturing of the upper limb that reproduced the figure four sign [4] at seizure onset. The left arm was extended, and the right arm was flexed (at the elbow), while the head was briefly turned to the left at this point of the seizure. Consequent bilateral clonic movements of the upper limbs showed a typical epileptic pattern with a gradual frequency decline and a progressive amplitude increase of movements [5]. A stereotyped left hemifacial spasm was also noted during the tonic contraction of the upper limbs. The short duration of the episodes, occasional evidence of cyanosis and lack of pupillary reactivity served as additional clues to the ES diagnosis. Although the clinical picture was initially confounded by the patient's preserved awareness during the bilateral convulsions, the absence of postictal confusion, and the normal EEG findings, these features finally appeared consistent with FBMS arising in the SSMA, as previously reported in the literature [[6], [7], [8]]. The patient was discharged with a normal neurologic status nine days after the last convulsive episode. Seizures did not recur even without antiseizure medication during the following month, after which the patient was lost to follow-up.Fig. 1 EEG recording of the initial phase of a focal bilateral motor seizure with a still-frame from the video.
EEG recording of the initial phase of a focal bilateral motor seizure with a still-frame from the video.
The EEG does not reveal evident abnormalities at seizure onset. The tracing is progressively disturbed by myogenic artifacts during the evolution of the seizure motor pattern.
Fig. 1Fig. 2 EEG recording of the final phase of a focal bilateral motor seizure.
Phasic myogenic artifacts mark the tracing during the ending clonic phase of the seizure. The EEG shows the persistence of the background activity in the immediate postictal period.
Fig. 2
3 Discussion
The manifestation of convulsive seizures without EEG abnormalities in the context of benzodiazepine withdrawal syndrome and psychiatric disease requires a differential diagnosis of ES and nonepileptic seizures (NES). In fact, the abrupt interruption of high dosage, chronic benzodiazepine treatment represents a potential precipitating factor for ES, while NES are occasionally observed as imitators of epilepsy in patients with mental health problems [9,10]. In addition to the psychiatric disorders and the normal neurophysiologic findings, the patient's preserved awareness during the bilateral convulsions and the absence of postictal confusion confounded the clinical picture with the potential to misdiagnose FBMS as NES. FBMS are described as motor onset seizures arising in one cerebral hemisphere and rapidly involving bilateral motor networks, often with asymmetric tonic posturing [11]. The SSMA is classically recognized as the anatomical origin of FBMS [11]. Clinical characteristics of focal onset seizures arising in the SSMA include short duration (typically not exceeding 40 s) [6,12]; predominant onset during sleep [6,8]; somatosensory aura [8,13]; preserved awareness [6,8,13]; absence of postictal confusion [8]; vocalization or speech arrest [6,8,13]; unilateral tonic posturing [8,13] or bilateral, asymmetric tonic posturing [4,13,14]; bilateral abduction of the upper extremities [6]; and bilateral thrashing movements [6,8] that usually constitute a paroxysmal, stereotyped motor pattern in the same individual [6]. Particularly, asymmetric tonic posturing has been related to the asynchronous activation of the SSMA [4,14], thus representing a key semiologic feature of seizures involving this cortical area [8,12]. Authors have also referred to asymmetric tonic posturing as the figure four sign that marks focal to bilateral tonic–clonic seizures [4]. The lateralizing value of this sign has been repeatedly reported with the epileptogenic focus usually contralateral to the arm that is extended at the elbow [4,14]. In contrast, NES with convulsive-like manifestations are characterized by long duration (frequently exceeding 2 min); fluctuating course; and asynchronous movements (typically side-to-side head movements and out of phase limb movements) that show unchanged frequency and variable amplitude throughout the event [10].
In our patient, the convulsive disorders shared most of the mentioned clinical characteristics of focal onset seizures arising in the SSMA. In particular, the origin of ictal symptoms in the SSMA was suggested by the early stereotyped contraction of the left arm and consequent asymmetric tonic posturing of limbs, while a left hemifacial spasm appeared consistent with the presumed epileptic focus in the right cerebral hemisphere. The progression of these phenomena to bilateral, tonic extension and clonic movements of the upper extremities characterized the final phase of the motor pattern mimicking focal to bilateral tonic–clonic seizures. However, the absence of awareness impairments, postictal confusion and EEG slowing in the immediate postictal phase discredited the apparent bilateral tonic–clonic evolution of focal seizures, suggesting limited epileptic involvement of the SSMA during the entire course of the events. While the seizure semiology could be referred to FBMS arising from the SSMA, the absence of ictal EEG abnormalities to confirm epileptic seizures is a limiting factor for the diagnosis of frontal lobe seizures. This suggests the potential utility of EEG recordings with additional midline electrodes (placed according to the 10–10 system) and/or ictal SPECT, as a supplementary method to confirm the epileptic nature and the anatomical origin of the seizures. However, it is not uncommon for seizures arising in the mesial and deep structures of the frontal lobe to have absent scalp ictal EEG changes utilizing the 10-20 international system of electrode placement.
Although structural brain abnormalities were not documented in our patient, neurochemical alterations induced by benzodiazepine withdrawal into specific networks may have resulted in a major seizure susceptibility of the SSMA. Interestingly, a major intrinsic epileptogenicity of specific cortical areas has been postulated by Albiero et al. to explain the occurrence of focal seizures in patients who underwent detoxification from chronic lormetazepam abuse [3]. Additional information is needed to ascertain the major occurrence of focal seizures in patients who interrupted chronic lormetazepam abuse. Although amitriptyline may have lowered the seizure threshold, the precise clinical impact of the substitution remains uncertain in our patient, as typical symptoms of amitriptyline intoxication such as sedation, arterial hypotension, cardiac arrhythmias or anticholinergic effects [15] were not observed during the period of recurrent seizures. In contrast, the strict concomitance of seizures with classic benzodiazepine abstinence symptoms such as insomnia, restlessness, rigidity, and muscular pain clearly indicated the primary precipitating role of lormetazepam withdrawal. A previous report remarked that lormetazepam is the benzodiazepine that is most frequently associated with abuse and dependence [16], particularly when it is used in the form of an oral solution [17]. The presence of ethanol in the oral solution of lormetazepam has been specifically suspected to increase the risks for abuse and dependence. Although ethanol quantities may not be relevant in patients treated with accurate doses of lormetazepam (1.5–2.5 mg daily for insomnia), the pathogenic effect of ethanol may be significant in lormetazepam addicted patients. The present case provides additional information about the risks of lormetazepam addiction as a component of psychiatric care. We suggest strict monitoring of patients treated with an oral solution of this medication.
4 Conclusion
FBMS may be observed during the clinical manifestation of lormetazepam withdrawal in patients who overdosed on amitriptyline. FBMS may produce a confusing clinical picture, with the potential to misdiagnose NES, especially when no abnormal neurophysiologic findings are clearly identified and when patients have a known underlying psychiatric disease.
The following is the supplementary data related to this article.Video 1
Focal bilateral motor seizure precipitated by lormetazepam withdrawal and amitriptyline overdose.
The seizure begins with a tonic posture of the left hand and abduction of the left arm flexed at the elbow. The left arm is then elevated and extended anteriorly, while the right arm is initially flexed at the elbow and rotated inward, reproducing the figure four sign. The head is briefly turned to the left during the asymmetric posturing of the upper extremities. Consequently, the right arm is also extended anteriorly. A slight facial cyanosis and a grimace with left hemifacial spasm (consistent with the presumed origin of the seizure in the right cerebral hemisphere) are noted during the extension of the upper limbs. There is no eye blink at the threatening gesture or pupillary response to the light at this point of the seizure. Convulsions cease with a sequence of vibratory and clonic movements of the upper extremities and trunk. The duration of the seizure is approximately 53 s and amitriptyline overdose.
Video 1
Ethical statement
The authors state that the study described in the paper complies with the publishing ethics of the journal.
The patient’s face has been partially obscured in the attached video, as requested by the patient herself at the time of the given consent.
Declaration of competing interest
The authors state that there is no conflict of interest. | Recovered | ReactionOutcome | CC BY-NC-ND | 33490945 | 18,821,337 | 2021 |
What was the outcome of reaction 'Drug withdrawal syndrome'? | Focal bilateral motor seizures precipitated by abrupt cessation of chronic lormetazepam abuse and amitriptyline overdose.
We report the case of an adult psychiatric patient who developed new-onset focal bilateral motor seizures (FBMS) in the context of a severe benzodiazepine withdrawal syndrome. The patient was forced to interrupt chronic lormetazepam abuse and overdosed on amitriptyline (800 mg in an oral solution) before the onset of seizures. Typical signs of amitriptyline intoxication such as sedation and anticholinergic effects were not observed. Video-EEG recordings revealed a stereotypical ictal motor pattern with asymmetric tonic posturing and bilateral clonic movements of the upper limbs, but there were no abnormalities identified by EEG. Seizures recurred multiple times per day but resolved simultaneously when withdrawal symptomatology subsided eight days after onset. Nonepileptic seizures (NES) were considered in the differential diagnosis because of the patient's psychiatric history including preserved awareness during the bilateral convulsions, the absence of postictal confusion, and normal EEG. The present case indicates that FBMS may occur during benzodiazepine withdrawal in patients who overdosed on amitriptyline. The diagnosis may be challenging as FBMS may mimic NES in the absence of abnormal neurophysiologic findings. This may be especially challenging in patients with an underlying psychiatric disease.
1 Introduction
Epileptic seizures (ES) have been frequently reported as clinical manifestations of benzodiazepine withdrawal, an acute disorder precipitated by an abrupt discontinuation of chronic therapies including benzodiazepines. The withdrawal symptomatology may also include insomnia, irritability, muscular pain and stiffness, hand tremor, anxiety, and psychotic episodes [1]. An acute reduction in brain gamma-aminobutyric acid function has been indicated as a pathogenic mechanism of the syndrome [2]. While generalized tonic–clonic seizures are the most common, focal seizures are rarely reported in this condition. Particularly, focal nonconvulsive seizures have been described in patients treated with flumazenil during detoxification from lormetazepam abuse [3].
2 Case report
A 46-year-old woman with a clinical history of schizoaffective and personality disorders presented to our hospital with new-onset convulsive seizures. It was reported that the patient was forced to interrupt chronic lormetazepam abuse (a daily drug consumption of 150–250 mg of an oral solution) the day before the onset of seizures. When she ran out of the medication, she ingested an oral solution containing 800 mg of amitriptyline as a supposed sedative substitute for lormetazepam. Convulsive seizures recurred multiple times per day in the hospital during daytime and night-time hours in a clinical context of insomnia, restlessness, muscular rigidity, and painful spasms affecting the back muscles that implicated a benzodiazepine withdrawal syndrome. There was no evidence of amitriptyline toxicity, as the patient did not show sedation, anticholinergic effects, arterial hypotension, or cardiac arrhythmias. The amitriptyline blood level was not obtained. Seizures presented with a duration of approximately 53 s and a paroxysmal, stereotyped sequence of movements that was documented with video-EEG. The recordings of seizures, using the usual 10–20 system of electrode placement, did not show EEG abnormalities or postictal slowing of the background activity, but the tracings were disturbed by myogenic artifacts throughout the course of the events (Fig. 1, Fig. 2). Awareness was preserved during the episodes. The patient occasionally called for help at seizure onset but remained unresponsive until the cessation of movements. Rapid recovery of speech and an absence of postictal confusion were reported on different occasions. Magnetic resonance imaging of the brain was normal. Three days after the patient's admission, a cluster of convulsive episodes prompted a short intensive treatment with propofol and mechanical ventilation. Seizures were increasingly observed again after extubation, despite oral treatment with clonazepam, levetiracetam, and repeated intravenous administrations of diazepam. Regardless of the inefficacy of the treatment during an eight-day period of intense seizure activity, the patient eventually became seizure-free. The remission of seizures occurred simultaneously with the cessation of insomnia and muscular symptoms related to lormetazepam withdrawal. Clonazepam and levetiracetam were subsequently reduced and discontinued as the patient remained asymptomatic. Despite the absence of ictal EEG abnormalities, a retrospective analysis of the seizure semiology based on video recordings and additional recorded observations of the episodes at bedside suggested the clinical diagnosis of FBMS, presumably arising from the supplementary sensorimotor area (SSMA). This diagnostic conclusion was supported by the evidence of a paroxysmal, stereotyped ictal motor pattern with early abduction of the arm flexed at the elbow and subsequent asymmetric tonic posturing of the upper limb that reproduced the figure four sign [4] at seizure onset. The left arm was extended, and the right arm was flexed (at the elbow), while the head was briefly turned to the left at this point of the seizure. Consequent bilateral clonic movements of the upper limbs showed a typical epileptic pattern with a gradual frequency decline and a progressive amplitude increase of movements [5]. A stereotyped left hemifacial spasm was also noted during the tonic contraction of the upper limbs. The short duration of the episodes, occasional evidence of cyanosis and lack of pupillary reactivity served as additional clues to the ES diagnosis. Although the clinical picture was initially confounded by the patient's preserved awareness during the bilateral convulsions, the absence of postictal confusion, and the normal EEG findings, these features finally appeared consistent with FBMS arising in the SSMA, as previously reported in the literature [[6], [7], [8]]. The patient was discharged with a normal neurologic status nine days after the last convulsive episode. Seizures did not recur even without antiseizure medication during the following month, after which the patient was lost to follow-up.Fig. 1 EEG recording of the initial phase of a focal bilateral motor seizure with a still-frame from the video.
EEG recording of the initial phase of a focal bilateral motor seizure with a still-frame from the video.
The EEG does not reveal evident abnormalities at seizure onset. The tracing is progressively disturbed by myogenic artifacts during the evolution of the seizure motor pattern.
Fig. 1Fig. 2 EEG recording of the final phase of a focal bilateral motor seizure.
Phasic myogenic artifacts mark the tracing during the ending clonic phase of the seizure. The EEG shows the persistence of the background activity in the immediate postictal period.
Fig. 2
3 Discussion
The manifestation of convulsive seizures without EEG abnormalities in the context of benzodiazepine withdrawal syndrome and psychiatric disease requires a differential diagnosis of ES and nonepileptic seizures (NES). In fact, the abrupt interruption of high dosage, chronic benzodiazepine treatment represents a potential precipitating factor for ES, while NES are occasionally observed as imitators of epilepsy in patients with mental health problems [9,10]. In addition to the psychiatric disorders and the normal neurophysiologic findings, the patient's preserved awareness during the bilateral convulsions and the absence of postictal confusion confounded the clinical picture with the potential to misdiagnose FBMS as NES. FBMS are described as motor onset seizures arising in one cerebral hemisphere and rapidly involving bilateral motor networks, often with asymmetric tonic posturing [11]. The SSMA is classically recognized as the anatomical origin of FBMS [11]. Clinical characteristics of focal onset seizures arising in the SSMA include short duration (typically not exceeding 40 s) [6,12]; predominant onset during sleep [6,8]; somatosensory aura [8,13]; preserved awareness [6,8,13]; absence of postictal confusion [8]; vocalization or speech arrest [6,8,13]; unilateral tonic posturing [8,13] or bilateral, asymmetric tonic posturing [4,13,14]; bilateral abduction of the upper extremities [6]; and bilateral thrashing movements [6,8] that usually constitute a paroxysmal, stereotyped motor pattern in the same individual [6]. Particularly, asymmetric tonic posturing has been related to the asynchronous activation of the SSMA [4,14], thus representing a key semiologic feature of seizures involving this cortical area [8,12]. Authors have also referred to asymmetric tonic posturing as the figure four sign that marks focal to bilateral tonic–clonic seizures [4]. The lateralizing value of this sign has been repeatedly reported with the epileptogenic focus usually contralateral to the arm that is extended at the elbow [4,14]. In contrast, NES with convulsive-like manifestations are characterized by long duration (frequently exceeding 2 min); fluctuating course; and asynchronous movements (typically side-to-side head movements and out of phase limb movements) that show unchanged frequency and variable amplitude throughout the event [10].
In our patient, the convulsive disorders shared most of the mentioned clinical characteristics of focal onset seizures arising in the SSMA. In particular, the origin of ictal symptoms in the SSMA was suggested by the early stereotyped contraction of the left arm and consequent asymmetric tonic posturing of limbs, while a left hemifacial spasm appeared consistent with the presumed epileptic focus in the right cerebral hemisphere. The progression of these phenomena to bilateral, tonic extension and clonic movements of the upper extremities characterized the final phase of the motor pattern mimicking focal to bilateral tonic–clonic seizures. However, the absence of awareness impairments, postictal confusion and EEG slowing in the immediate postictal phase discredited the apparent bilateral tonic–clonic evolution of focal seizures, suggesting limited epileptic involvement of the SSMA during the entire course of the events. While the seizure semiology could be referred to FBMS arising from the SSMA, the absence of ictal EEG abnormalities to confirm epileptic seizures is a limiting factor for the diagnosis of frontal lobe seizures. This suggests the potential utility of EEG recordings with additional midline electrodes (placed according to the 10–10 system) and/or ictal SPECT, as a supplementary method to confirm the epileptic nature and the anatomical origin of the seizures. However, it is not uncommon for seizures arising in the mesial and deep structures of the frontal lobe to have absent scalp ictal EEG changes utilizing the 10-20 international system of electrode placement.
Although structural brain abnormalities were not documented in our patient, neurochemical alterations induced by benzodiazepine withdrawal into specific networks may have resulted in a major seizure susceptibility of the SSMA. Interestingly, a major intrinsic epileptogenicity of specific cortical areas has been postulated by Albiero et al. to explain the occurrence of focal seizures in patients who underwent detoxification from chronic lormetazepam abuse [3]. Additional information is needed to ascertain the major occurrence of focal seizures in patients who interrupted chronic lormetazepam abuse. Although amitriptyline may have lowered the seizure threshold, the precise clinical impact of the substitution remains uncertain in our patient, as typical symptoms of amitriptyline intoxication such as sedation, arterial hypotension, cardiac arrhythmias or anticholinergic effects [15] were not observed during the period of recurrent seizures. In contrast, the strict concomitance of seizures with classic benzodiazepine abstinence symptoms such as insomnia, restlessness, rigidity, and muscular pain clearly indicated the primary precipitating role of lormetazepam withdrawal. A previous report remarked that lormetazepam is the benzodiazepine that is most frequently associated with abuse and dependence [16], particularly when it is used in the form of an oral solution [17]. The presence of ethanol in the oral solution of lormetazepam has been specifically suspected to increase the risks for abuse and dependence. Although ethanol quantities may not be relevant in patients treated with accurate doses of lormetazepam (1.5–2.5 mg daily for insomnia), the pathogenic effect of ethanol may be significant in lormetazepam addicted patients. The present case provides additional information about the risks of lormetazepam addiction as a component of psychiatric care. We suggest strict monitoring of patients treated with an oral solution of this medication.
4 Conclusion
FBMS may be observed during the clinical manifestation of lormetazepam withdrawal in patients who overdosed on amitriptyline. FBMS may produce a confusing clinical picture, with the potential to misdiagnose NES, especially when no abnormal neurophysiologic findings are clearly identified and when patients have a known underlying psychiatric disease.
The following is the supplementary data related to this article.Video 1
Focal bilateral motor seizure precipitated by lormetazepam withdrawal and amitriptyline overdose.
The seizure begins with a tonic posture of the left hand and abduction of the left arm flexed at the elbow. The left arm is then elevated and extended anteriorly, while the right arm is initially flexed at the elbow and rotated inward, reproducing the figure four sign. The head is briefly turned to the left during the asymmetric posturing of the upper extremities. Consequently, the right arm is also extended anteriorly. A slight facial cyanosis and a grimace with left hemifacial spasm (consistent with the presumed origin of the seizure in the right cerebral hemisphere) are noted during the extension of the upper limbs. There is no eye blink at the threatening gesture or pupillary response to the light at this point of the seizure. Convulsions cease with a sequence of vibratory and clonic movements of the upper extremities and trunk. The duration of the seizure is approximately 53 s and amitriptyline overdose.
Video 1
Ethical statement
The authors state that the study described in the paper complies with the publishing ethics of the journal.
The patient’s face has been partially obscured in the attached video, as requested by the patient herself at the time of the given consent.
Declaration of competing interest
The authors state that there is no conflict of interest. | Recovered | ReactionOutcome | CC BY-NC-ND | 33490945 | 18,821,337 | 2021 |
What was the outcome of reaction 'Intentional product misuse'? | Focal bilateral motor seizures precipitated by abrupt cessation of chronic lormetazepam abuse and amitriptyline overdose.
We report the case of an adult psychiatric patient who developed new-onset focal bilateral motor seizures (FBMS) in the context of a severe benzodiazepine withdrawal syndrome. The patient was forced to interrupt chronic lormetazepam abuse and overdosed on amitriptyline (800 mg in an oral solution) before the onset of seizures. Typical signs of amitriptyline intoxication such as sedation and anticholinergic effects were not observed. Video-EEG recordings revealed a stereotypical ictal motor pattern with asymmetric tonic posturing and bilateral clonic movements of the upper limbs, but there were no abnormalities identified by EEG. Seizures recurred multiple times per day but resolved simultaneously when withdrawal symptomatology subsided eight days after onset. Nonepileptic seizures (NES) were considered in the differential diagnosis because of the patient's psychiatric history including preserved awareness during the bilateral convulsions, the absence of postictal confusion, and normal EEG. The present case indicates that FBMS may occur during benzodiazepine withdrawal in patients who overdosed on amitriptyline. The diagnosis may be challenging as FBMS may mimic NES in the absence of abnormal neurophysiologic findings. This may be especially challenging in patients with an underlying psychiatric disease.
1 Introduction
Epileptic seizures (ES) have been frequently reported as clinical manifestations of benzodiazepine withdrawal, an acute disorder precipitated by an abrupt discontinuation of chronic therapies including benzodiazepines. The withdrawal symptomatology may also include insomnia, irritability, muscular pain and stiffness, hand tremor, anxiety, and psychotic episodes [1]. An acute reduction in brain gamma-aminobutyric acid function has been indicated as a pathogenic mechanism of the syndrome [2]. While generalized tonic–clonic seizures are the most common, focal seizures are rarely reported in this condition. Particularly, focal nonconvulsive seizures have been described in patients treated with flumazenil during detoxification from lormetazepam abuse [3].
2 Case report
A 46-year-old woman with a clinical history of schizoaffective and personality disorders presented to our hospital with new-onset convulsive seizures. It was reported that the patient was forced to interrupt chronic lormetazepam abuse (a daily drug consumption of 150–250 mg of an oral solution) the day before the onset of seizures. When she ran out of the medication, she ingested an oral solution containing 800 mg of amitriptyline as a supposed sedative substitute for lormetazepam. Convulsive seizures recurred multiple times per day in the hospital during daytime and night-time hours in a clinical context of insomnia, restlessness, muscular rigidity, and painful spasms affecting the back muscles that implicated a benzodiazepine withdrawal syndrome. There was no evidence of amitriptyline toxicity, as the patient did not show sedation, anticholinergic effects, arterial hypotension, or cardiac arrhythmias. The amitriptyline blood level was not obtained. Seizures presented with a duration of approximately 53 s and a paroxysmal, stereotyped sequence of movements that was documented with video-EEG. The recordings of seizures, using the usual 10–20 system of electrode placement, did not show EEG abnormalities or postictal slowing of the background activity, but the tracings were disturbed by myogenic artifacts throughout the course of the events (Fig. 1, Fig. 2). Awareness was preserved during the episodes. The patient occasionally called for help at seizure onset but remained unresponsive until the cessation of movements. Rapid recovery of speech and an absence of postictal confusion were reported on different occasions. Magnetic resonance imaging of the brain was normal. Three days after the patient's admission, a cluster of convulsive episodes prompted a short intensive treatment with propofol and mechanical ventilation. Seizures were increasingly observed again after extubation, despite oral treatment with clonazepam, levetiracetam, and repeated intravenous administrations of diazepam. Regardless of the inefficacy of the treatment during an eight-day period of intense seizure activity, the patient eventually became seizure-free. The remission of seizures occurred simultaneously with the cessation of insomnia and muscular symptoms related to lormetazepam withdrawal. Clonazepam and levetiracetam were subsequently reduced and discontinued as the patient remained asymptomatic. Despite the absence of ictal EEG abnormalities, a retrospective analysis of the seizure semiology based on video recordings and additional recorded observations of the episodes at bedside suggested the clinical diagnosis of FBMS, presumably arising from the supplementary sensorimotor area (SSMA). This diagnostic conclusion was supported by the evidence of a paroxysmal, stereotyped ictal motor pattern with early abduction of the arm flexed at the elbow and subsequent asymmetric tonic posturing of the upper limb that reproduced the figure four sign [4] at seizure onset. The left arm was extended, and the right arm was flexed (at the elbow), while the head was briefly turned to the left at this point of the seizure. Consequent bilateral clonic movements of the upper limbs showed a typical epileptic pattern with a gradual frequency decline and a progressive amplitude increase of movements [5]. A stereotyped left hemifacial spasm was also noted during the tonic contraction of the upper limbs. The short duration of the episodes, occasional evidence of cyanosis and lack of pupillary reactivity served as additional clues to the ES diagnosis. Although the clinical picture was initially confounded by the patient's preserved awareness during the bilateral convulsions, the absence of postictal confusion, and the normal EEG findings, these features finally appeared consistent with FBMS arising in the SSMA, as previously reported in the literature [[6], [7], [8]]. The patient was discharged with a normal neurologic status nine days after the last convulsive episode. Seizures did not recur even without antiseizure medication during the following month, after which the patient was lost to follow-up.Fig. 1 EEG recording of the initial phase of a focal bilateral motor seizure with a still-frame from the video.
EEG recording of the initial phase of a focal bilateral motor seizure with a still-frame from the video.
The EEG does not reveal evident abnormalities at seizure onset. The tracing is progressively disturbed by myogenic artifacts during the evolution of the seizure motor pattern.
Fig. 1Fig. 2 EEG recording of the final phase of a focal bilateral motor seizure.
Phasic myogenic artifacts mark the tracing during the ending clonic phase of the seizure. The EEG shows the persistence of the background activity in the immediate postictal period.
Fig. 2
3 Discussion
The manifestation of convulsive seizures without EEG abnormalities in the context of benzodiazepine withdrawal syndrome and psychiatric disease requires a differential diagnosis of ES and nonepileptic seizures (NES). In fact, the abrupt interruption of high dosage, chronic benzodiazepine treatment represents a potential precipitating factor for ES, while NES are occasionally observed as imitators of epilepsy in patients with mental health problems [9,10]. In addition to the psychiatric disorders and the normal neurophysiologic findings, the patient's preserved awareness during the bilateral convulsions and the absence of postictal confusion confounded the clinical picture with the potential to misdiagnose FBMS as NES. FBMS are described as motor onset seizures arising in one cerebral hemisphere and rapidly involving bilateral motor networks, often with asymmetric tonic posturing [11]. The SSMA is classically recognized as the anatomical origin of FBMS [11]. Clinical characteristics of focal onset seizures arising in the SSMA include short duration (typically not exceeding 40 s) [6,12]; predominant onset during sleep [6,8]; somatosensory aura [8,13]; preserved awareness [6,8,13]; absence of postictal confusion [8]; vocalization or speech arrest [6,8,13]; unilateral tonic posturing [8,13] or bilateral, asymmetric tonic posturing [4,13,14]; bilateral abduction of the upper extremities [6]; and bilateral thrashing movements [6,8] that usually constitute a paroxysmal, stereotyped motor pattern in the same individual [6]. Particularly, asymmetric tonic posturing has been related to the asynchronous activation of the SSMA [4,14], thus representing a key semiologic feature of seizures involving this cortical area [8,12]. Authors have also referred to asymmetric tonic posturing as the figure four sign that marks focal to bilateral tonic–clonic seizures [4]. The lateralizing value of this sign has been repeatedly reported with the epileptogenic focus usually contralateral to the arm that is extended at the elbow [4,14]. In contrast, NES with convulsive-like manifestations are characterized by long duration (frequently exceeding 2 min); fluctuating course; and asynchronous movements (typically side-to-side head movements and out of phase limb movements) that show unchanged frequency and variable amplitude throughout the event [10].
In our patient, the convulsive disorders shared most of the mentioned clinical characteristics of focal onset seizures arising in the SSMA. In particular, the origin of ictal symptoms in the SSMA was suggested by the early stereotyped contraction of the left arm and consequent asymmetric tonic posturing of limbs, while a left hemifacial spasm appeared consistent with the presumed epileptic focus in the right cerebral hemisphere. The progression of these phenomena to bilateral, tonic extension and clonic movements of the upper extremities characterized the final phase of the motor pattern mimicking focal to bilateral tonic–clonic seizures. However, the absence of awareness impairments, postictal confusion and EEG slowing in the immediate postictal phase discredited the apparent bilateral tonic–clonic evolution of focal seizures, suggesting limited epileptic involvement of the SSMA during the entire course of the events. While the seizure semiology could be referred to FBMS arising from the SSMA, the absence of ictal EEG abnormalities to confirm epileptic seizures is a limiting factor for the diagnosis of frontal lobe seizures. This suggests the potential utility of EEG recordings with additional midline electrodes (placed according to the 10–10 system) and/or ictal SPECT, as a supplementary method to confirm the epileptic nature and the anatomical origin of the seizures. However, it is not uncommon for seizures arising in the mesial and deep structures of the frontal lobe to have absent scalp ictal EEG changes utilizing the 10-20 international system of electrode placement.
Although structural brain abnormalities were not documented in our patient, neurochemical alterations induced by benzodiazepine withdrawal into specific networks may have resulted in a major seizure susceptibility of the SSMA. Interestingly, a major intrinsic epileptogenicity of specific cortical areas has been postulated by Albiero et al. to explain the occurrence of focal seizures in patients who underwent detoxification from chronic lormetazepam abuse [3]. Additional information is needed to ascertain the major occurrence of focal seizures in patients who interrupted chronic lormetazepam abuse. Although amitriptyline may have lowered the seizure threshold, the precise clinical impact of the substitution remains uncertain in our patient, as typical symptoms of amitriptyline intoxication such as sedation, arterial hypotension, cardiac arrhythmias or anticholinergic effects [15] were not observed during the period of recurrent seizures. In contrast, the strict concomitance of seizures with classic benzodiazepine abstinence symptoms such as insomnia, restlessness, rigidity, and muscular pain clearly indicated the primary precipitating role of lormetazepam withdrawal. A previous report remarked that lormetazepam is the benzodiazepine that is most frequently associated with abuse and dependence [16], particularly when it is used in the form of an oral solution [17]. The presence of ethanol in the oral solution of lormetazepam has been specifically suspected to increase the risks for abuse and dependence. Although ethanol quantities may not be relevant in patients treated with accurate doses of lormetazepam (1.5–2.5 mg daily for insomnia), the pathogenic effect of ethanol may be significant in lormetazepam addicted patients. The present case provides additional information about the risks of lormetazepam addiction as a component of psychiatric care. We suggest strict monitoring of patients treated with an oral solution of this medication.
4 Conclusion
FBMS may be observed during the clinical manifestation of lormetazepam withdrawal in patients who overdosed on amitriptyline. FBMS may produce a confusing clinical picture, with the potential to misdiagnose NES, especially when no abnormal neurophysiologic findings are clearly identified and when patients have a known underlying psychiatric disease.
The following is the supplementary data related to this article.Video 1
Focal bilateral motor seizure precipitated by lormetazepam withdrawal and amitriptyline overdose.
The seizure begins with a tonic posture of the left hand and abduction of the left arm flexed at the elbow. The left arm is then elevated and extended anteriorly, while the right arm is initially flexed at the elbow and rotated inward, reproducing the figure four sign. The head is briefly turned to the left during the asymmetric posturing of the upper extremities. Consequently, the right arm is also extended anteriorly. A slight facial cyanosis and a grimace with left hemifacial spasm (consistent with the presumed origin of the seizure in the right cerebral hemisphere) are noted during the extension of the upper limbs. There is no eye blink at the threatening gesture or pupillary response to the light at this point of the seizure. Convulsions cease with a sequence of vibratory and clonic movements of the upper extremities and trunk. The duration of the seizure is approximately 53 s and amitriptyline overdose.
Video 1
Ethical statement
The authors state that the study described in the paper complies with the publishing ethics of the journal.
The patient’s face has been partially obscured in the attached video, as requested by the patient herself at the time of the given consent.
Declaration of competing interest
The authors state that there is no conflict of interest. | Recovered | ReactionOutcome | CC BY-NC-ND | 33490945 | 18,821,337 | 2021 |
What was the outcome of reaction 'Overdose'? | Focal bilateral motor seizures precipitated by abrupt cessation of chronic lormetazepam abuse and amitriptyline overdose.
We report the case of an adult psychiatric patient who developed new-onset focal bilateral motor seizures (FBMS) in the context of a severe benzodiazepine withdrawal syndrome. The patient was forced to interrupt chronic lormetazepam abuse and overdosed on amitriptyline (800 mg in an oral solution) before the onset of seizures. Typical signs of amitriptyline intoxication such as sedation and anticholinergic effects were not observed. Video-EEG recordings revealed a stereotypical ictal motor pattern with asymmetric tonic posturing and bilateral clonic movements of the upper limbs, but there were no abnormalities identified by EEG. Seizures recurred multiple times per day but resolved simultaneously when withdrawal symptomatology subsided eight days after onset. Nonepileptic seizures (NES) were considered in the differential diagnosis because of the patient's psychiatric history including preserved awareness during the bilateral convulsions, the absence of postictal confusion, and normal EEG. The present case indicates that FBMS may occur during benzodiazepine withdrawal in patients who overdosed on amitriptyline. The diagnosis may be challenging as FBMS may mimic NES in the absence of abnormal neurophysiologic findings. This may be especially challenging in patients with an underlying psychiatric disease.
1 Introduction
Epileptic seizures (ES) have been frequently reported as clinical manifestations of benzodiazepine withdrawal, an acute disorder precipitated by an abrupt discontinuation of chronic therapies including benzodiazepines. The withdrawal symptomatology may also include insomnia, irritability, muscular pain and stiffness, hand tremor, anxiety, and psychotic episodes [1]. An acute reduction in brain gamma-aminobutyric acid function has been indicated as a pathogenic mechanism of the syndrome [2]. While generalized tonic–clonic seizures are the most common, focal seizures are rarely reported in this condition. Particularly, focal nonconvulsive seizures have been described in patients treated with flumazenil during detoxification from lormetazepam abuse [3].
2 Case report
A 46-year-old woman with a clinical history of schizoaffective and personality disorders presented to our hospital with new-onset convulsive seizures. It was reported that the patient was forced to interrupt chronic lormetazepam abuse (a daily drug consumption of 150–250 mg of an oral solution) the day before the onset of seizures. When she ran out of the medication, she ingested an oral solution containing 800 mg of amitriptyline as a supposed sedative substitute for lormetazepam. Convulsive seizures recurred multiple times per day in the hospital during daytime and night-time hours in a clinical context of insomnia, restlessness, muscular rigidity, and painful spasms affecting the back muscles that implicated a benzodiazepine withdrawal syndrome. There was no evidence of amitriptyline toxicity, as the patient did not show sedation, anticholinergic effects, arterial hypotension, or cardiac arrhythmias. The amitriptyline blood level was not obtained. Seizures presented with a duration of approximately 53 s and a paroxysmal, stereotyped sequence of movements that was documented with video-EEG. The recordings of seizures, using the usual 10–20 system of electrode placement, did not show EEG abnormalities or postictal slowing of the background activity, but the tracings were disturbed by myogenic artifacts throughout the course of the events (Fig. 1, Fig. 2). Awareness was preserved during the episodes. The patient occasionally called for help at seizure onset but remained unresponsive until the cessation of movements. Rapid recovery of speech and an absence of postictal confusion were reported on different occasions. Magnetic resonance imaging of the brain was normal. Three days after the patient's admission, a cluster of convulsive episodes prompted a short intensive treatment with propofol and mechanical ventilation. Seizures were increasingly observed again after extubation, despite oral treatment with clonazepam, levetiracetam, and repeated intravenous administrations of diazepam. Regardless of the inefficacy of the treatment during an eight-day period of intense seizure activity, the patient eventually became seizure-free. The remission of seizures occurred simultaneously with the cessation of insomnia and muscular symptoms related to lormetazepam withdrawal. Clonazepam and levetiracetam were subsequently reduced and discontinued as the patient remained asymptomatic. Despite the absence of ictal EEG abnormalities, a retrospective analysis of the seizure semiology based on video recordings and additional recorded observations of the episodes at bedside suggested the clinical diagnosis of FBMS, presumably arising from the supplementary sensorimotor area (SSMA). This diagnostic conclusion was supported by the evidence of a paroxysmal, stereotyped ictal motor pattern with early abduction of the arm flexed at the elbow and subsequent asymmetric tonic posturing of the upper limb that reproduced the figure four sign [4] at seizure onset. The left arm was extended, and the right arm was flexed (at the elbow), while the head was briefly turned to the left at this point of the seizure. Consequent bilateral clonic movements of the upper limbs showed a typical epileptic pattern with a gradual frequency decline and a progressive amplitude increase of movements [5]. A stereotyped left hemifacial spasm was also noted during the tonic contraction of the upper limbs. The short duration of the episodes, occasional evidence of cyanosis and lack of pupillary reactivity served as additional clues to the ES diagnosis. Although the clinical picture was initially confounded by the patient's preserved awareness during the bilateral convulsions, the absence of postictal confusion, and the normal EEG findings, these features finally appeared consistent with FBMS arising in the SSMA, as previously reported in the literature [[6], [7], [8]]. The patient was discharged with a normal neurologic status nine days after the last convulsive episode. Seizures did not recur even without antiseizure medication during the following month, after which the patient was lost to follow-up.Fig. 1 EEG recording of the initial phase of a focal bilateral motor seizure with a still-frame from the video.
EEG recording of the initial phase of a focal bilateral motor seizure with a still-frame from the video.
The EEG does not reveal evident abnormalities at seizure onset. The tracing is progressively disturbed by myogenic artifacts during the evolution of the seizure motor pattern.
Fig. 1Fig. 2 EEG recording of the final phase of a focal bilateral motor seizure.
Phasic myogenic artifacts mark the tracing during the ending clonic phase of the seizure. The EEG shows the persistence of the background activity in the immediate postictal period.
Fig. 2
3 Discussion
The manifestation of convulsive seizures without EEG abnormalities in the context of benzodiazepine withdrawal syndrome and psychiatric disease requires a differential diagnosis of ES and nonepileptic seizures (NES). In fact, the abrupt interruption of high dosage, chronic benzodiazepine treatment represents a potential precipitating factor for ES, while NES are occasionally observed as imitators of epilepsy in patients with mental health problems [9,10]. In addition to the psychiatric disorders and the normal neurophysiologic findings, the patient's preserved awareness during the bilateral convulsions and the absence of postictal confusion confounded the clinical picture with the potential to misdiagnose FBMS as NES. FBMS are described as motor onset seizures arising in one cerebral hemisphere and rapidly involving bilateral motor networks, often with asymmetric tonic posturing [11]. The SSMA is classically recognized as the anatomical origin of FBMS [11]. Clinical characteristics of focal onset seizures arising in the SSMA include short duration (typically not exceeding 40 s) [6,12]; predominant onset during sleep [6,8]; somatosensory aura [8,13]; preserved awareness [6,8,13]; absence of postictal confusion [8]; vocalization or speech arrest [6,8,13]; unilateral tonic posturing [8,13] or bilateral, asymmetric tonic posturing [4,13,14]; bilateral abduction of the upper extremities [6]; and bilateral thrashing movements [6,8] that usually constitute a paroxysmal, stereotyped motor pattern in the same individual [6]. Particularly, asymmetric tonic posturing has been related to the asynchronous activation of the SSMA [4,14], thus representing a key semiologic feature of seizures involving this cortical area [8,12]. Authors have also referred to asymmetric tonic posturing as the figure four sign that marks focal to bilateral tonic–clonic seizures [4]. The lateralizing value of this sign has been repeatedly reported with the epileptogenic focus usually contralateral to the arm that is extended at the elbow [4,14]. In contrast, NES with convulsive-like manifestations are characterized by long duration (frequently exceeding 2 min); fluctuating course; and asynchronous movements (typically side-to-side head movements and out of phase limb movements) that show unchanged frequency and variable amplitude throughout the event [10].
In our patient, the convulsive disorders shared most of the mentioned clinical characteristics of focal onset seizures arising in the SSMA. In particular, the origin of ictal symptoms in the SSMA was suggested by the early stereotyped contraction of the left arm and consequent asymmetric tonic posturing of limbs, while a left hemifacial spasm appeared consistent with the presumed epileptic focus in the right cerebral hemisphere. The progression of these phenomena to bilateral, tonic extension and clonic movements of the upper extremities characterized the final phase of the motor pattern mimicking focal to bilateral tonic–clonic seizures. However, the absence of awareness impairments, postictal confusion and EEG slowing in the immediate postictal phase discredited the apparent bilateral tonic–clonic evolution of focal seizures, suggesting limited epileptic involvement of the SSMA during the entire course of the events. While the seizure semiology could be referred to FBMS arising from the SSMA, the absence of ictal EEG abnormalities to confirm epileptic seizures is a limiting factor for the diagnosis of frontal lobe seizures. This suggests the potential utility of EEG recordings with additional midline electrodes (placed according to the 10–10 system) and/or ictal SPECT, as a supplementary method to confirm the epileptic nature and the anatomical origin of the seizures. However, it is not uncommon for seizures arising in the mesial and deep structures of the frontal lobe to have absent scalp ictal EEG changes utilizing the 10-20 international system of electrode placement.
Although structural brain abnormalities were not documented in our patient, neurochemical alterations induced by benzodiazepine withdrawal into specific networks may have resulted in a major seizure susceptibility of the SSMA. Interestingly, a major intrinsic epileptogenicity of specific cortical areas has been postulated by Albiero et al. to explain the occurrence of focal seizures in patients who underwent detoxification from chronic lormetazepam abuse [3]. Additional information is needed to ascertain the major occurrence of focal seizures in patients who interrupted chronic lormetazepam abuse. Although amitriptyline may have lowered the seizure threshold, the precise clinical impact of the substitution remains uncertain in our patient, as typical symptoms of amitriptyline intoxication such as sedation, arterial hypotension, cardiac arrhythmias or anticholinergic effects [15] were not observed during the period of recurrent seizures. In contrast, the strict concomitance of seizures with classic benzodiazepine abstinence symptoms such as insomnia, restlessness, rigidity, and muscular pain clearly indicated the primary precipitating role of lormetazepam withdrawal. A previous report remarked that lormetazepam is the benzodiazepine that is most frequently associated with abuse and dependence [16], particularly when it is used in the form of an oral solution [17]. The presence of ethanol in the oral solution of lormetazepam has been specifically suspected to increase the risks for abuse and dependence. Although ethanol quantities may not be relevant in patients treated with accurate doses of lormetazepam (1.5–2.5 mg daily for insomnia), the pathogenic effect of ethanol may be significant in lormetazepam addicted patients. The present case provides additional information about the risks of lormetazepam addiction as a component of psychiatric care. We suggest strict monitoring of patients treated with an oral solution of this medication.
4 Conclusion
FBMS may be observed during the clinical manifestation of lormetazepam withdrawal in patients who overdosed on amitriptyline. FBMS may produce a confusing clinical picture, with the potential to misdiagnose NES, especially when no abnormal neurophysiologic findings are clearly identified and when patients have a known underlying psychiatric disease.
The following is the supplementary data related to this article.Video 1
Focal bilateral motor seizure precipitated by lormetazepam withdrawal and amitriptyline overdose.
The seizure begins with a tonic posture of the left hand and abduction of the left arm flexed at the elbow. The left arm is then elevated and extended anteriorly, while the right arm is initially flexed at the elbow and rotated inward, reproducing the figure four sign. The head is briefly turned to the left during the asymmetric posturing of the upper extremities. Consequently, the right arm is also extended anteriorly. A slight facial cyanosis and a grimace with left hemifacial spasm (consistent with the presumed origin of the seizure in the right cerebral hemisphere) are noted during the extension of the upper limbs. There is no eye blink at the threatening gesture or pupillary response to the light at this point of the seizure. Convulsions cease with a sequence of vibratory and clonic movements of the upper extremities and trunk. The duration of the seizure is approximately 53 s and amitriptyline overdose.
Video 1
Ethical statement
The authors state that the study described in the paper complies with the publishing ethics of the journal.
The patient’s face has been partially obscured in the attached video, as requested by the patient herself at the time of the given consent.
Declaration of competing interest
The authors state that there is no conflict of interest. | Recovered | ReactionOutcome | CC BY-NC-ND | 33490945 | 18,821,337 | 2021 |
What was the outcome of reaction 'Partial seizures'? | Focal bilateral motor seizures precipitated by abrupt cessation of chronic lormetazepam abuse and amitriptyline overdose.
We report the case of an adult psychiatric patient who developed new-onset focal bilateral motor seizures (FBMS) in the context of a severe benzodiazepine withdrawal syndrome. The patient was forced to interrupt chronic lormetazepam abuse and overdosed on amitriptyline (800 mg in an oral solution) before the onset of seizures. Typical signs of amitriptyline intoxication such as sedation and anticholinergic effects were not observed. Video-EEG recordings revealed a stereotypical ictal motor pattern with asymmetric tonic posturing and bilateral clonic movements of the upper limbs, but there were no abnormalities identified by EEG. Seizures recurred multiple times per day but resolved simultaneously when withdrawal symptomatology subsided eight days after onset. Nonepileptic seizures (NES) were considered in the differential diagnosis because of the patient's psychiatric history including preserved awareness during the bilateral convulsions, the absence of postictal confusion, and normal EEG. The present case indicates that FBMS may occur during benzodiazepine withdrawal in patients who overdosed on amitriptyline. The diagnosis may be challenging as FBMS may mimic NES in the absence of abnormal neurophysiologic findings. This may be especially challenging in patients with an underlying psychiatric disease.
1 Introduction
Epileptic seizures (ES) have been frequently reported as clinical manifestations of benzodiazepine withdrawal, an acute disorder precipitated by an abrupt discontinuation of chronic therapies including benzodiazepines. The withdrawal symptomatology may also include insomnia, irritability, muscular pain and stiffness, hand tremor, anxiety, and psychotic episodes [1]. An acute reduction in brain gamma-aminobutyric acid function has been indicated as a pathogenic mechanism of the syndrome [2]. While generalized tonic–clonic seizures are the most common, focal seizures are rarely reported in this condition. Particularly, focal nonconvulsive seizures have been described in patients treated with flumazenil during detoxification from lormetazepam abuse [3].
2 Case report
A 46-year-old woman with a clinical history of schizoaffective and personality disorders presented to our hospital with new-onset convulsive seizures. It was reported that the patient was forced to interrupt chronic lormetazepam abuse (a daily drug consumption of 150–250 mg of an oral solution) the day before the onset of seizures. When she ran out of the medication, she ingested an oral solution containing 800 mg of amitriptyline as a supposed sedative substitute for lormetazepam. Convulsive seizures recurred multiple times per day in the hospital during daytime and night-time hours in a clinical context of insomnia, restlessness, muscular rigidity, and painful spasms affecting the back muscles that implicated a benzodiazepine withdrawal syndrome. There was no evidence of amitriptyline toxicity, as the patient did not show sedation, anticholinergic effects, arterial hypotension, or cardiac arrhythmias. The amitriptyline blood level was not obtained. Seizures presented with a duration of approximately 53 s and a paroxysmal, stereotyped sequence of movements that was documented with video-EEG. The recordings of seizures, using the usual 10–20 system of electrode placement, did not show EEG abnormalities or postictal slowing of the background activity, but the tracings were disturbed by myogenic artifacts throughout the course of the events (Fig. 1, Fig. 2). Awareness was preserved during the episodes. The patient occasionally called for help at seizure onset but remained unresponsive until the cessation of movements. Rapid recovery of speech and an absence of postictal confusion were reported on different occasions. Magnetic resonance imaging of the brain was normal. Three days after the patient's admission, a cluster of convulsive episodes prompted a short intensive treatment with propofol and mechanical ventilation. Seizures were increasingly observed again after extubation, despite oral treatment with clonazepam, levetiracetam, and repeated intravenous administrations of diazepam. Regardless of the inefficacy of the treatment during an eight-day period of intense seizure activity, the patient eventually became seizure-free. The remission of seizures occurred simultaneously with the cessation of insomnia and muscular symptoms related to lormetazepam withdrawal. Clonazepam and levetiracetam were subsequently reduced and discontinued as the patient remained asymptomatic. Despite the absence of ictal EEG abnormalities, a retrospective analysis of the seizure semiology based on video recordings and additional recorded observations of the episodes at bedside suggested the clinical diagnosis of FBMS, presumably arising from the supplementary sensorimotor area (SSMA). This diagnostic conclusion was supported by the evidence of a paroxysmal, stereotyped ictal motor pattern with early abduction of the arm flexed at the elbow and subsequent asymmetric tonic posturing of the upper limb that reproduced the figure four sign [4] at seizure onset. The left arm was extended, and the right arm was flexed (at the elbow), while the head was briefly turned to the left at this point of the seizure. Consequent bilateral clonic movements of the upper limbs showed a typical epileptic pattern with a gradual frequency decline and a progressive amplitude increase of movements [5]. A stereotyped left hemifacial spasm was also noted during the tonic contraction of the upper limbs. The short duration of the episodes, occasional evidence of cyanosis and lack of pupillary reactivity served as additional clues to the ES diagnosis. Although the clinical picture was initially confounded by the patient's preserved awareness during the bilateral convulsions, the absence of postictal confusion, and the normal EEG findings, these features finally appeared consistent with FBMS arising in the SSMA, as previously reported in the literature [[6], [7], [8]]. The patient was discharged with a normal neurologic status nine days after the last convulsive episode. Seizures did not recur even without antiseizure medication during the following month, after which the patient was lost to follow-up.Fig. 1 EEG recording of the initial phase of a focal bilateral motor seizure with a still-frame from the video.
EEG recording of the initial phase of a focal bilateral motor seizure with a still-frame from the video.
The EEG does not reveal evident abnormalities at seizure onset. The tracing is progressively disturbed by myogenic artifacts during the evolution of the seizure motor pattern.
Fig. 1Fig. 2 EEG recording of the final phase of a focal bilateral motor seizure.
Phasic myogenic artifacts mark the tracing during the ending clonic phase of the seizure. The EEG shows the persistence of the background activity in the immediate postictal period.
Fig. 2
3 Discussion
The manifestation of convulsive seizures without EEG abnormalities in the context of benzodiazepine withdrawal syndrome and psychiatric disease requires a differential diagnosis of ES and nonepileptic seizures (NES). In fact, the abrupt interruption of high dosage, chronic benzodiazepine treatment represents a potential precipitating factor for ES, while NES are occasionally observed as imitators of epilepsy in patients with mental health problems [9,10]. In addition to the psychiatric disorders and the normal neurophysiologic findings, the patient's preserved awareness during the bilateral convulsions and the absence of postictal confusion confounded the clinical picture with the potential to misdiagnose FBMS as NES. FBMS are described as motor onset seizures arising in one cerebral hemisphere and rapidly involving bilateral motor networks, often with asymmetric tonic posturing [11]. The SSMA is classically recognized as the anatomical origin of FBMS [11]. Clinical characteristics of focal onset seizures arising in the SSMA include short duration (typically not exceeding 40 s) [6,12]; predominant onset during sleep [6,8]; somatosensory aura [8,13]; preserved awareness [6,8,13]; absence of postictal confusion [8]; vocalization or speech arrest [6,8,13]; unilateral tonic posturing [8,13] or bilateral, asymmetric tonic posturing [4,13,14]; bilateral abduction of the upper extremities [6]; and bilateral thrashing movements [6,8] that usually constitute a paroxysmal, stereotyped motor pattern in the same individual [6]. Particularly, asymmetric tonic posturing has been related to the asynchronous activation of the SSMA [4,14], thus representing a key semiologic feature of seizures involving this cortical area [8,12]. Authors have also referred to asymmetric tonic posturing as the figure four sign that marks focal to bilateral tonic–clonic seizures [4]. The lateralizing value of this sign has been repeatedly reported with the epileptogenic focus usually contralateral to the arm that is extended at the elbow [4,14]. In contrast, NES with convulsive-like manifestations are characterized by long duration (frequently exceeding 2 min); fluctuating course; and asynchronous movements (typically side-to-side head movements and out of phase limb movements) that show unchanged frequency and variable amplitude throughout the event [10].
In our patient, the convulsive disorders shared most of the mentioned clinical characteristics of focal onset seizures arising in the SSMA. In particular, the origin of ictal symptoms in the SSMA was suggested by the early stereotyped contraction of the left arm and consequent asymmetric tonic posturing of limbs, while a left hemifacial spasm appeared consistent with the presumed epileptic focus in the right cerebral hemisphere. The progression of these phenomena to bilateral, tonic extension and clonic movements of the upper extremities characterized the final phase of the motor pattern mimicking focal to bilateral tonic–clonic seizures. However, the absence of awareness impairments, postictal confusion and EEG slowing in the immediate postictal phase discredited the apparent bilateral tonic–clonic evolution of focal seizures, suggesting limited epileptic involvement of the SSMA during the entire course of the events. While the seizure semiology could be referred to FBMS arising from the SSMA, the absence of ictal EEG abnormalities to confirm epileptic seizures is a limiting factor for the diagnosis of frontal lobe seizures. This suggests the potential utility of EEG recordings with additional midline electrodes (placed according to the 10–10 system) and/or ictal SPECT, as a supplementary method to confirm the epileptic nature and the anatomical origin of the seizures. However, it is not uncommon for seizures arising in the mesial and deep structures of the frontal lobe to have absent scalp ictal EEG changes utilizing the 10-20 international system of electrode placement.
Although structural brain abnormalities were not documented in our patient, neurochemical alterations induced by benzodiazepine withdrawal into specific networks may have resulted in a major seizure susceptibility of the SSMA. Interestingly, a major intrinsic epileptogenicity of specific cortical areas has been postulated by Albiero et al. to explain the occurrence of focal seizures in patients who underwent detoxification from chronic lormetazepam abuse [3]. Additional information is needed to ascertain the major occurrence of focal seizures in patients who interrupted chronic lormetazepam abuse. Although amitriptyline may have lowered the seizure threshold, the precise clinical impact of the substitution remains uncertain in our patient, as typical symptoms of amitriptyline intoxication such as sedation, arterial hypotension, cardiac arrhythmias or anticholinergic effects [15] were not observed during the period of recurrent seizures. In contrast, the strict concomitance of seizures with classic benzodiazepine abstinence symptoms such as insomnia, restlessness, rigidity, and muscular pain clearly indicated the primary precipitating role of lormetazepam withdrawal. A previous report remarked that lormetazepam is the benzodiazepine that is most frequently associated with abuse and dependence [16], particularly when it is used in the form of an oral solution [17]. The presence of ethanol in the oral solution of lormetazepam has been specifically suspected to increase the risks for abuse and dependence. Although ethanol quantities may not be relevant in patients treated with accurate doses of lormetazepam (1.5–2.5 mg daily for insomnia), the pathogenic effect of ethanol may be significant in lormetazepam addicted patients. The present case provides additional information about the risks of lormetazepam addiction as a component of psychiatric care. We suggest strict monitoring of patients treated with an oral solution of this medication.
4 Conclusion
FBMS may be observed during the clinical manifestation of lormetazepam withdrawal in patients who overdosed on amitriptyline. FBMS may produce a confusing clinical picture, with the potential to misdiagnose NES, especially when no abnormal neurophysiologic findings are clearly identified and when patients have a known underlying psychiatric disease.
The following is the supplementary data related to this article.Video 1
Focal bilateral motor seizure precipitated by lormetazepam withdrawal and amitriptyline overdose.
The seizure begins with a tonic posture of the left hand and abduction of the left arm flexed at the elbow. The left arm is then elevated and extended anteriorly, while the right arm is initially flexed at the elbow and rotated inward, reproducing the figure four sign. The head is briefly turned to the left during the asymmetric posturing of the upper extremities. Consequently, the right arm is also extended anteriorly. A slight facial cyanosis and a grimace with left hemifacial spasm (consistent with the presumed origin of the seizure in the right cerebral hemisphere) are noted during the extension of the upper limbs. There is no eye blink at the threatening gesture or pupillary response to the light at this point of the seizure. Convulsions cease with a sequence of vibratory and clonic movements of the upper extremities and trunk. The duration of the seizure is approximately 53 s and amitriptyline overdose.
Video 1
Ethical statement
The authors state that the study described in the paper complies with the publishing ethics of the journal.
The patient’s face has been partially obscured in the attached video, as requested by the patient herself at the time of the given consent.
Declaration of competing interest
The authors state that there is no conflict of interest. | Recovered | ReactionOutcome | CC BY-NC-ND | 33490945 | 18,821,337 | 2021 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Haematoma muscle'. | Spontaneous ilio-psoas haematomas (IPHs): a warning for COVID-19 inpatients.
Critically ill patients with COVID-19 are at increased risk of developing a hypercoagulable state due to haemostatic changes directly related to the SARS-CoV-2 infection or to the consequence of the cytokine storm. Anticoagulation is now recommended to reduce the thrombotic risk. Ilio-psoas haematoma (IPH) is a potentially lethal condition that can arise during the hospitalization, especially in intensive care units (ICUs) and frequently reported as a complication of anticoagulation treatment.
We report a case series of seven subjects with SARS-CoV-2 pneumonia complicated by Ilio-psoas haematomas (IPHs) at our COVID-Hospital in Rome, Italy.
Over the observation period, 925 subjects with confirmed SARS-CoV-2 infection were admitted to our COVID-hospital. Among them, we found seven spontaneous IPHs with an incidence of 7.6 cases per 1000 hospitalization. All the reported cases had a severe manifestation of COVID-19 pneumonia, with at least one comorbidity and 5/7 were on treatment with low weight molecular heparin for micro or macro pulmonary thrombosis.
Given the indications to prescribe anticoagulant therapy in COVID-19 and the lack of solid evidences on the optimal dose and duration, it is important to be aware of the iliopsoas haematoma as a potentially serious complication in COVID-19 inpatients. KEY MESSAGE Critically ill patients with COVID-19 are at increased risk of hypercoagulability state and anticoagulation therapy is recommended. Ilio-psoas haematoma (IPH) is found to be a complication of anticoagulation regimen especially in severe COVID-19 cases. An incidence of 7.6 cases per 1000 admission of IPHs was reported. Hypoesthesia of the lower limbs, pain triggered by femoral rotation, hypovolaemia and anaemia are the most common symptoms and signs of IPHs that should alert physician.
Introduction
Critically ill patients with coronavirus disease-2019 (COVID-19) are at increased risk of developing a hypercoagulable state [1].
The pathophysiology behind this phenomenon has been suspected to be a result of haemostatic changes that might be direct effect of severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2) or a consequence of a cytokine storm that alters the onset of the systemic inflammatory response syndrome (SIRS), as observed in other viral disease [2]. To hinder this hypercoagulable state, COVID-19 patients should be properly anticoagulated to reduce the thrombotic risk [3–8].
IPH is a potentially lethal condition frequently reported as a complication of anticoagulation therapy that can arise during the hospitalization, especially in intensive care units (ICUs) [9]. To date, a single case of spontaneous IPH in a COVID-19 patient has been reported [10]. Here, we report seven cases of spontaneous IPHs occurring in patients with severe COVID-19 pneumonia admitted to our Institute.
Materials and methods
Study population and setting
We consecutively included all subjects with a microbiologically confirmed SARS-CoV-2 infection who were admitted to the National Institute for Infectious Diseases IRCCS Lazzaro Spallanzani in Rome, Italy, between 1 March 2020 and 30 October 2020. Medical history, demographic and clinical data were collected through review of medical records. Data have been collected for the ReCOVeRI Study, a registry on COVID-19 for clinical Research of the National Institute for Infectious Diseases L. Spallanzani, approved by the Ethical Committee of the National Institute for Infectious Diseases L. Spallanzani IRCCS (number 164, 26 June 2020). Laboratory and radiologic assessments during the hospital stay were performed by the treating physician according to the hospital operative procedure. All patients gave informed consent for collecting personal data for research purposes.
Definitions
A confirmed case of COVID-19 was defined by a positive real-time reverse-transcription PCR (RT-PCR) assay for SARS-CoV-2 on nasopharyngeal swab and/or a positive serology for SARS-CoV-2 (positive immunoglobulin (Ig) G or M or A for SARS-CoV-2).
Severe disease was defined as clinical signs of pneumonia plus one of the following: respiratory rate greater than 30 breaths per min, severe respiratory distress or oxygen saturation less than 90% on room air [11].
Hyperinflammation syndrome was defined as having at least two among D-dimer above 1000 ng/mL, ferritin above 500 mcg/L, LDH above 300 UI/L and lymphocyte count below 1000 cell/mm3 [1]. Ilio-psoas haematoma was diagnosed by using computerized tomography (CT)-scan or magnetic resonance imaging (MRI).
All CT scans were performed on a multi-detector row CT scanner (Bright Speed, General Electric Medical Systems, Milwaukee, WI) using 120 kV pp, 250 mA, pitch of 1.375, gantry rotation time of 0.6 s. A chest and abdomen CT was performed from the apex of lung to symphysis pubis before and after injection of iodinated contrast media into a peripheral vein with three-phase arterial, venous and delayed phase. The baseline scan of the thorax was reconstructed with slice thicknesses of 0.625 mm and spacing of 1 mm with high contrast resolution algorithm. The contrast media scan of the thorax and abdomen was reconstructed with slice thicknesses of 1.25 mm and spacing of 1 mm, completed with multiplanar reconstructions (MPR and Mip). The MRI scan was performed on HDx scanner (General Electric Medical Systems, Milwaukee, WI), 1.5 T, with T1 and T2 weighted sequences also after spectral fat subtraction and paramagnetic contrast media.
Results
Over the observation period, 925 subjects with confirmed SARS-CoV-2 infection were admitted to our COVID-hospital. Among them, we observed seven spontaneous IPHs with an incidence of 7.6 cases per 1000 hospitalization.
Clinical characteristics and outcome
Main demographic and clinical characteristics and outcomes are summarized in Table 1.
Table 1. Characteristics of the reported cases.
Patient 1 Patient 2 Patient 3 Patient 4 Patient 5 Patient 6 Patient 7
Age 80 75 94 55 64 72 66
Gender M F F M F M F
Comorbidities Atrial fibrillation; COPD Hypertension; asthma Diabetes, ischaemic heart disease, hypertension, cognitive impairment, bedridden syndrome Not reported Ischaemic heart disease, obesity Diabetes, hypertension Thyroid disease
Clinical presentation at admission
Days from symptoms onset to admission 14 5 2 14 7 7 7
Symptoms Arthralgia, asthenia, dyspnoea, fever, cough Headache, dyspnoea, fever, cough, myalgia Fever, cough Fever, cough Fever, cough Fever, cough, dyspnoea Fever, dysgeusia, cough
SpO2 at admission on room air 90% 96% 95% 93% 91% 90% 92%
Days from admission to IPH 9 25 14 5 38 30 18
Days from admission to prophylactic anticoagulation start 1 9 0 0 0 0 0
Length of hospital stay (days) 20 50 36 34 – 70 29
Exitus Died Discharged Discharged Discharged Still admitted Discharged Discharged
Inflammatory Index at admission
Haemoglobin, g/dL 13.9 11.6 12.8 7.4 11 12.3 13.8
Platelets, ×103/mm3 121 142 185 176 255 193 350
Lymphocytes, cell ×103/mm3 4.2 9.6 12.2 15.0 8.1 7.5 12.21
LDH, UI/L 404 576 254 249 325 205 433
D-dimer, ng/mL 796 740 551 1373 608 10,850 901
Fibrinogen, mg/dL 798 487 47 314 824 531 706
ferritin, ng/mL 2167 4754 1113 548 303 369 534
PT, INR 0.93 1.08 1.23 1.18 1.06 1.12 1.21
PTT, sec 48.4 35.7 48.7 33.1 24.6 26.3 26.6
C-reactive protein, md/dL 15.78 8.62 33.1 16.2 12.9 4.5 6.6
Inflammatory index at ilio-psoas haematoma
Haemoglobin, g/dL 8.7 6.8 10.5 8 6.9 4.9 8.8
Platelet, ×103/mm3 234 91 295 173 492 217 410
Lymphocytes, cell ×103/mm3 6.3 10.3 13.8 9.4 16.1 13.4 11.6
LDH, UI/L 377 441 254 208 410 – 417
D-dimer, ng/mL 796 2033 1287 1517 728 8000 NA
Fibrinogen, mg/dL 351 241 685 361 382 450
Ferritin, ng/mL 2167 3506 903 1010 1000 NA NA
PT, INR 1.06 1.13 0.94 1.33 1.13 1.18
aPTT, s 29.3 16.6 38.6 33.9 40.7 28.2
C-reactive protein, md/dL 0.62 1.52 1.97 12.23 2.28 13 4.73
Treatment during hospitalization
HCQ No Yes No No No No No
Lopinavir/ritonavir Yes Yes Yes No No No No
Remdesivir – – Yes – Yes Yes Yes
Steroids Methyl-prednisolone Methyl-prednisolone No No No Desametasone Desametasone
LMWH prophylaxis No Yes Yes Yes Yes Yes Yes
LMWH treatment 8000 UI ×2/day No No 6000 UI ×2/day 6000 UI ×2/day 6000 UI ×2/day 6000 UI ×2/day
Need for oxygen supplement Yes (cPAP) Yes (cPAP) Yes (VM with FiO2 40%) Yes (OTI) Yes (OTI) Yes (NIV) Yes (OTI)
IPH: ilio-psoas haematoma; LDH: lactic dehydrogenase; NA: not available; PT: prothrombin time; INR: international normalized ratio; aPTT: activated partial thromboplastin time; HCQ: hydroxychloroquine; LMWH: low molecular weight heparin; cPAP: continuous positive airway pressure; VM: venturi mask; OTI: orotracheal intubation; NIV: non-invasive ventilation.
Four patients were female with an age ranging between 65 and 80 years and a median body mass index of 28 (IQR 25–32). At the time of COVID-19 diagnosis, all patients were diagnosed with pneumonia and had at least one comorbidity; among them, three patients had hypertension, two patients had diabetes and one patient a chronic obstructive pulmonary disease (COPD). The median hospital stay was 34 days (IQR 30–36). Severe clinical presentation was observed in four patients with an admission oxygen saturation (SpO2) 93% on room air (IQR 90–98). The arterial oxygen partial pressure (PaO2 in mmHg) to fractional inspired oxygen (PaO2/FiO2) ratio was between 200 mmHg and 300 mmHg and it required supplemental oxygen therapy with non-invasive ventilation by using continuous positive airway pressure (cPAP). Three patients required invasive mechanical ventilation with orotracheal intubation (OTI) and admission to the ICU. Chest CT scan performed at admission, showed bilateral ground-glass opacities (GGOs) and sub-segmental consolidations, mostly located in the peripheral zone. Moreover, other imaging features such as linear opacities, “crazy-paving” pattern, the “reverse halo sign” and subsegmental vessel enlargement were described. In particular, the vessel enlargement was described close to the GGOs, which is compatible with thrombo-inflammatory processes. The contrast enhancement scan showed in five out of seven thrombosis-mediated micro-perfusion defects in peripheral pulmonary vessels.
All patients received prophylactic low weight molecular heparin (LWMH) at admission, with the exception of one patient who started anticoagulant therapy with LWMH for pulmonary embolism previously diagnosed at the emergency department. The heparin dosages were modified according to clinical worsening and to the diagnosis of micro or macro pulmonary thromboembolic events. Only one patient was taking antiaggregants medications discontinued as soon as the diagnosis of IPH was made.
Overall, steroid therapy was administered in five patients. Specifically, concerning SARS-CoV-2 therapy, four patients received oral lopinavir/ritonavir (LPV/r, 400/100 mg twice per day for 14 days), one of them received also oral hydroxychloroquine (200 mg twice per day for 10 days); four patients were treated with intravenous remdesivir (200 mg on day one followed by 100 mg since day 2 to day 10), combined with intravenous dexamethasone (6 mg once daily for 10 days) in three of them. Inflammation and coagulation parameters are shown in Table 1. Briefly, all patients had an hyperinflammation pattern with a median ferritin level of 548 pg/mL (IQR 369–2167 pg/mL) and C-reactive protein (CRP) of 8.6 mg/dL (IQR 6.6–15.8 mg/dL) and a D-dimer of 796 ng/mL (IQR 608–1373 ng/mL).
During the hospitalization, after a median of 35 (IQR 29–50) days, signs of neurologic compression with hypoesthesia of the lower limbs and general signs of hypovolaemia and anaemia were always present. A CT scan of the abdomen was performed revealing iliopsoas haematoma in all patients apart from one subject studied by performing an MRI.
The patients were treated in a conservative manner, resuscitated, according to clinical judgement, with intravenous fluid, transfusions of red blood cells and other supportive measures; only one patient, haemodynamically unstable, underwent arterial embolization.
Five patients were discharged with no long-lasting complications, one patient is still hospitalized and one patient died.
Imaging
The differential diagnosis of pelvic mass of the abdominal wall includes the most frequent pathologies as sarcoma, haematoma and abscess. Mass signal characteristics, vascular pattern during and after intravenous injection of contrast media and the presence of other elements (e.g. small intralesional calcifications) might help the radiologist to perform a correct diagnosis.
In our cases, a non-enhanced CT scan of the abdomen showed diffuse enlargement as well as heterogeneous density of the ilio-psoas muscle with an area of high density or fluid-fluid level for subacute haematoma. The absence of a contiguous mass with a vascular pattern led to rule out a diagnosis of haemorrhagic sarcoma.
The size of the haematomas found ranged from a minimum of 4–10 cm. In four patients, the dimensions exceeded 12 cm of transverse diameter and longitudinal extension of about 10 cm.
In two cases, the CT angiogram showed multiple streaks of contrast blush; delayed phase shows pooling (increased size of blush) of intravenous contrast media (Figure 1(a,b)). Contrast media blush, lack of internal fluid density and marginal enhancing component may exclude the diagnosis of iliopsoas abscess and raises the possibility of subacute haematoma.
Figure 1. (a) Arterial phase CT of right haematoma and (b) delayed phase CT of right haematoma.
In one case, MRI study was performed and showed a heterogeneous iliopsoas mass, with normal diffusion-weighted imaging (DWI) pattern and hyperintense areas on T1- and T2-weighted images for haemoglobin catabolites (metaHb) for subacute haematoma (Figure 2(a–c)).
Figure 2. (a) MRI of right iliopsoas haematoma, (b) MRI of right iliopsoas haematoma and (c) MRI of right iliopsoas haematoma.
The MRI findings correlated with the evolution of the haematoma:Acute haematoma: Iso-hypointense in T1-weighted (w) or slightly hypointense to muscle and hypo-hyperintense in T2 weighted (w).
Subacute haematoma: Hyperintense in T1w and T2w; high intensity rim, higher intensity peripheral zone and lower intensity core in T1w and relatively higher signal from core to periphery in T2w.
Chronic haematoma: Hypointense rim in T1w and T2w.
Discussion
Spontaneous IPH is defined as a retroperitoneal collection of blood involving the ilio-psoas muscle. Few studies evaluated the incidence of spontaneous IPHs in patients undergoing anticoagulation therapy that has been reported ranging from 0.1 to 0.6% [12]. Recently, a retrospective study in no COVID patients showed an incidence of IPHs of 3.8 cases per 1000 admission in ICUs [9] and our incidence of 7.6 cases over 1000 hospitalization, is higher than that previously reported. Risk factors in our small case series are the same reported in literature, as age, anticoagulation, a high body mass index, comorbidities such as hypertension and diabetes [9,13,14], moreover, our patients had an activated partial thromboplastin time (aPTT) above the therapeutic range at the moment of the diagnosis of the IPHs. Disseminated intravascular coagulopathy (DIC) and increased aPTT are both independent predictors of unfavourable prognosis [9]. The precise pathogenesis of retroperitoneal bleeding is unknown, it is most commonly reported as a complication of anticoagulation and, more rarely, in the setting of a clotting disorder or traumatic injury during the patients’ mobilization in the prone position [15–19].
Due to the anatomical proximity of these muscles to the lumbar plexus, it is hypothesized that retroperitoneal pre-existing microvascular atherosclerosis could increase sensitivity to rupture and microtrauma such as cough or vomiting could also lead to retroperitoneal bleeding [20].
It is well known that one of the causes of mortality in COVID-19 patients is venous thromboembolism (VTE) as evidenced by altered coagulation profile like elevated D-dimers [1,20].
This hypercoagulable phenomenon is due to the increased pro-inflammatory cytokines leading to atherosclerotic changes through local inflammation, microvascular thrombi and haemodynamic changes with multiorgan failure and death [1]. For these reasons, the administration of prophylactic or therapeutic anticoagulant agents is recommended [3–8]. Five cases among those we reported were fully treated with low molecular weight heparin for pulmonary micro-thrombosis.
Anticoagulation treatment and additional anti-platelet medications, increase the risk of major bleeding complications like retroperitoneal haemorrhage [9,10]. In a French ICU, 19 out of 92 (21%) COVID-19 patients on full anticoagulant treatment had 22 haemorrhagic events, and five of them were gastrointestinal [21].
Regarding IPHs, the optimal treatment remains controversial; however, initial treatment consists of discontinuation of anticoagulant agents, transfusion therapy, volume resuscitation and supportive measures; haemodynamically unstable patients, according to expert clinical judgement, can be treated with arterial embolization as it is minimally invasive with quick therapeutic effect when compared with surgical treatment.
Furthermore, stopping anticoagulation in case of IPHs and of a documented pulmonary micro-thrombosis in COVID-19 could be life-threatening.
Given the indications to prescribe anticoagulation in COVID-19 and the lack of solid evidences on the optimal dose and duration, specifically in micro-thrombosis, it is important to be aware of the iliopsoas haematoma as a potentially serious complication. As there is no consensus on therapeutical management of IPHs, each decision (i.e. conservative treatment, embolization, surgical or CT scan-guided haematoma’s drainage) should be made according to the clinical stability of the patients and by weighting risks and benefits.
We definitely need more studies in order to establish which is the optimal heparin dose and how long it should be continued in micro-thrombosis, as anticoagulation may increase the risk of major and potentially fatal bleeding.
Disclosure statement
All the authors have no competing interests that might be perceived to influence the results and/or discussion reported in this paper. Outside of this submitted work: Alessandra Vergori received institutional grant from Gilead Sciences, personal fees and travel grant from Janssen, personal fee from MSD; Andrea Antinori has served as a paid consultant to Gilead Sciences, Janssen-Cilag, Merck and ViiV Healthcare and received research institutional grants from Gilead Sciences, Janssen-Cilag and ViiV Healthcare. The other co-authors declare no conflicts of interests outside the submitted work.
Data availability statement
Data available on request from the authors.
Appendix The authors gratefully acknowledge nurse staff, all the patients and all members of the ReCOVeRI Study Group: Maria Alessandra Abbonizio, Amina Abdeddaim, Elisabetta Agostini, Chiara Agrati, Fabrizio Albarello, Gioia Amadei, Alessandra Amendola, Andrea Antinori, Maria Assunta Antonica, Mario Antonini, Tommaso Ascoli Bartoli, Francesco Baldini, Raffaella Barbaro, Barbara Bartolini, Rita Bellagamba, Martina Benigni, Nazario Bevilacqua, Gianluigi Biava, Michele Bibas, Licia Bordi, Veronica Bordoni, Evangelo Boumis, Marta Branca, Rosanna Buonomo, Donatella Busso, Marta Camici, Paolo Campioni, Flaminia Canichella, Maria Rosaria Capobianchi, Alessandro Capone, Cinzia Caporale, Emanuela Caraffa, Ilaria Caravella, Fabrizio Carletti, Concetta Castilletti, Adriana Cataldo, Stefano Cerilli, Carlotta Cerva, Roberta Chiappini, Pierangelo Chinello, Maria Assunta Cianfarani, Carmine Ciaralli, Claudia Cimaglia, Nicola Cinicola, Veronica Ciotti, Stefania Cicalini, Francesca Colavita, Angela Corpolongo, Massimo Cristofaro, Salvatore Curiale, Alessandra D’Abramo, Cristina Dantimi, Alessia De Angelis, Giada De Angelis, Maria Grazia De Palo, Federico De Zottis, Virginia Di Bari, Rachele Di Lorenzo, Federica Di Stefano, Gianpiero D’Offizi, Davide Donno, Francesca Evangelista, Francesca Faraglia, Anna Farina, Federica Ferraro, Lorena Fiorentini, Andrea Frustaci, Matteo Fusetti, Vincenzo Galati, Roberta Gagliardini, Paola Gallì, Gabriele Garotto, Ilaria Gaviano, Saba Gebremeskel Tekle, Maria Letizia Giancola, Filippo Giansante, Emanuela Giombini, Guido Granata, Maria Cristina Greci, Elisabetta Grilli, Susanna Grisetti, Gina Gualano, Fabio Iacomi, Marta Iaconi, Giuseppina Iannicelli, Carlo Inversi, Giuseppe Ippolito, Eleonora Lalle, Maria Elena Lamanna, Simone Lanini, Daniele Lapa, Luciana Lepore, Raffaella Libertone, Raffaella Lionetti, Giuseppina Liuzzi, Laura Loiacono, Andrea Lucia, Franco Lufrani, Manuela Macchione, Gaetano Maffongelli, Alessandra Marani, Luisa Marchioni, Andrea Mariano, Maria Cristina Marini, Micaela Maritti, Annelisa Mastrobattista, Ilaria Mastrorosa, Giulia Matusali, Valentina Mazzotta, Paola Mencarini, Silvia Meschi, Francesco Messina, Sibiana Micarelli, Giulia Mogavero, Annalisa Mondi, Marzia Montalbano, Chiara Montaldo, Silvia Mosti, Silvia Murachelli, Maria Musso, Michela Nardi, Assunta Navarra, Emanuele Nicastri, Martina Nocioni, Pasquale Noto, Roberto Noto, Alessandra Oliva, Ilaria Onnis, Sandrine Ottou, Claudia Palazzolo, Emanuele Pallini, Fabrizio Palmieri, Giulio Palombi, Carlo Pareo, Virgilio Passeri, Federico Pelliccioni, Giovanna Penna, Antonella Petrecchia, Ada Petrone, Nicola Petrosillo, Elisa Pianura, Carmela Pinnetti, Maria Pisciotta, Pierluca Piselli, Silvia Pittalis, Agostina Pontarelli, Costanza Proietti, Vincenzo Puro, Paolo Migliorisi Ramazzini, Alessia Rianda, Gabriele Rinonapoli, Silvia Rosati, Dorotea Rubino, Martina Rueca, Alberto Ruggeri, Alessandra Sacchi, Alessandro Sampaolesi, Francesco Sanasi, Carmen Santagata, Alessandra Scarabello, Silvana Scarcia, Vincenzo Schininà, Paola Scognamiglio, Laura Scorzolini, Giulia Stazi, Giacomo Strano,Fabrizio Taglietti, Chiara Taibi, Giorgia Taloni, Tetaj Nardi, Roberto Tonnarini, Simone Topino, Martina Tozzi, Francesco Vaia, Francesco Vairo, Maria Beatrice Valli, Alessandra Vergori, Laura Vincenzi, Ubaldo Visco-Comandini, Serena Vita, Pietro Vittozzi, Mauro Zaccarelli, Antonella Zanetti and Sara Zito. | HYDROXYCHLOROQUINE, LOPINAVIR\RITONAVIR, METHYLPREDNISOLONE, UNSPECIFIED INGREDIENT | DrugsGivenReaction | CC BY | 33491498 | 19,026,066 | 2021-12 |
What was the administration route of drug 'LOPINAVIR\RITONAVIR'? | Spontaneous ilio-psoas haematomas (IPHs): a warning for COVID-19 inpatients.
Critically ill patients with COVID-19 are at increased risk of developing a hypercoagulable state due to haemostatic changes directly related to the SARS-CoV-2 infection or to the consequence of the cytokine storm. Anticoagulation is now recommended to reduce the thrombotic risk. Ilio-psoas haematoma (IPH) is a potentially lethal condition that can arise during the hospitalization, especially in intensive care units (ICUs) and frequently reported as a complication of anticoagulation treatment.
We report a case series of seven subjects with SARS-CoV-2 pneumonia complicated by Ilio-psoas haematomas (IPHs) at our COVID-Hospital in Rome, Italy.
Over the observation period, 925 subjects with confirmed SARS-CoV-2 infection were admitted to our COVID-hospital. Among them, we found seven spontaneous IPHs with an incidence of 7.6 cases per 1000 hospitalization. All the reported cases had a severe manifestation of COVID-19 pneumonia, with at least one comorbidity and 5/7 were on treatment with low weight molecular heparin for micro or macro pulmonary thrombosis.
Given the indications to prescribe anticoagulant therapy in COVID-19 and the lack of solid evidences on the optimal dose and duration, it is important to be aware of the iliopsoas haematoma as a potentially serious complication in COVID-19 inpatients. KEY MESSAGE Critically ill patients with COVID-19 are at increased risk of hypercoagulability state and anticoagulation therapy is recommended. Ilio-psoas haematoma (IPH) is found to be a complication of anticoagulation regimen especially in severe COVID-19 cases. An incidence of 7.6 cases per 1000 admission of IPHs was reported. Hypoesthesia of the lower limbs, pain triggered by femoral rotation, hypovolaemia and anaemia are the most common symptoms and signs of IPHs that should alert physician.
Introduction
Critically ill patients with coronavirus disease-2019 (COVID-19) are at increased risk of developing a hypercoagulable state [1].
The pathophysiology behind this phenomenon has been suspected to be a result of haemostatic changes that might be direct effect of severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2) or a consequence of a cytokine storm that alters the onset of the systemic inflammatory response syndrome (SIRS), as observed in other viral disease [2]. To hinder this hypercoagulable state, COVID-19 patients should be properly anticoagulated to reduce the thrombotic risk [3–8].
IPH is a potentially lethal condition frequently reported as a complication of anticoagulation therapy that can arise during the hospitalization, especially in intensive care units (ICUs) [9]. To date, a single case of spontaneous IPH in a COVID-19 patient has been reported [10]. Here, we report seven cases of spontaneous IPHs occurring in patients with severe COVID-19 pneumonia admitted to our Institute.
Materials and methods
Study population and setting
We consecutively included all subjects with a microbiologically confirmed SARS-CoV-2 infection who were admitted to the National Institute for Infectious Diseases IRCCS Lazzaro Spallanzani in Rome, Italy, between 1 March 2020 and 30 October 2020. Medical history, demographic and clinical data were collected through review of medical records. Data have been collected for the ReCOVeRI Study, a registry on COVID-19 for clinical Research of the National Institute for Infectious Diseases L. Spallanzani, approved by the Ethical Committee of the National Institute for Infectious Diseases L. Spallanzani IRCCS (number 164, 26 June 2020). Laboratory and radiologic assessments during the hospital stay were performed by the treating physician according to the hospital operative procedure. All patients gave informed consent for collecting personal data for research purposes.
Definitions
A confirmed case of COVID-19 was defined by a positive real-time reverse-transcription PCR (RT-PCR) assay for SARS-CoV-2 on nasopharyngeal swab and/or a positive serology for SARS-CoV-2 (positive immunoglobulin (Ig) G or M or A for SARS-CoV-2).
Severe disease was defined as clinical signs of pneumonia plus one of the following: respiratory rate greater than 30 breaths per min, severe respiratory distress or oxygen saturation less than 90% on room air [11].
Hyperinflammation syndrome was defined as having at least two among D-dimer above 1000 ng/mL, ferritin above 500 mcg/L, LDH above 300 UI/L and lymphocyte count below 1000 cell/mm3 [1]. Ilio-psoas haematoma was diagnosed by using computerized tomography (CT)-scan or magnetic resonance imaging (MRI).
All CT scans were performed on a multi-detector row CT scanner (Bright Speed, General Electric Medical Systems, Milwaukee, WI) using 120 kV pp, 250 mA, pitch of 1.375, gantry rotation time of 0.6 s. A chest and abdomen CT was performed from the apex of lung to symphysis pubis before and after injection of iodinated contrast media into a peripheral vein with three-phase arterial, venous and delayed phase. The baseline scan of the thorax was reconstructed with slice thicknesses of 0.625 mm and spacing of 1 mm with high contrast resolution algorithm. The contrast media scan of the thorax and abdomen was reconstructed with slice thicknesses of 1.25 mm and spacing of 1 mm, completed with multiplanar reconstructions (MPR and Mip). The MRI scan was performed on HDx scanner (General Electric Medical Systems, Milwaukee, WI), 1.5 T, with T1 and T2 weighted sequences also after spectral fat subtraction and paramagnetic contrast media.
Results
Over the observation period, 925 subjects with confirmed SARS-CoV-2 infection were admitted to our COVID-hospital. Among them, we observed seven spontaneous IPHs with an incidence of 7.6 cases per 1000 hospitalization.
Clinical characteristics and outcome
Main demographic and clinical characteristics and outcomes are summarized in Table 1.
Table 1. Characteristics of the reported cases.
Patient 1 Patient 2 Patient 3 Patient 4 Patient 5 Patient 6 Patient 7
Age 80 75 94 55 64 72 66
Gender M F F M F M F
Comorbidities Atrial fibrillation; COPD Hypertension; asthma Diabetes, ischaemic heart disease, hypertension, cognitive impairment, bedridden syndrome Not reported Ischaemic heart disease, obesity Diabetes, hypertension Thyroid disease
Clinical presentation at admission
Days from symptoms onset to admission 14 5 2 14 7 7 7
Symptoms Arthralgia, asthenia, dyspnoea, fever, cough Headache, dyspnoea, fever, cough, myalgia Fever, cough Fever, cough Fever, cough Fever, cough, dyspnoea Fever, dysgeusia, cough
SpO2 at admission on room air 90% 96% 95% 93% 91% 90% 92%
Days from admission to IPH 9 25 14 5 38 30 18
Days from admission to prophylactic anticoagulation start 1 9 0 0 0 0 0
Length of hospital stay (days) 20 50 36 34 – 70 29
Exitus Died Discharged Discharged Discharged Still admitted Discharged Discharged
Inflammatory Index at admission
Haemoglobin, g/dL 13.9 11.6 12.8 7.4 11 12.3 13.8
Platelets, ×103/mm3 121 142 185 176 255 193 350
Lymphocytes, cell ×103/mm3 4.2 9.6 12.2 15.0 8.1 7.5 12.21
LDH, UI/L 404 576 254 249 325 205 433
D-dimer, ng/mL 796 740 551 1373 608 10,850 901
Fibrinogen, mg/dL 798 487 47 314 824 531 706
ferritin, ng/mL 2167 4754 1113 548 303 369 534
PT, INR 0.93 1.08 1.23 1.18 1.06 1.12 1.21
PTT, sec 48.4 35.7 48.7 33.1 24.6 26.3 26.6
C-reactive protein, md/dL 15.78 8.62 33.1 16.2 12.9 4.5 6.6
Inflammatory index at ilio-psoas haematoma
Haemoglobin, g/dL 8.7 6.8 10.5 8 6.9 4.9 8.8
Platelet, ×103/mm3 234 91 295 173 492 217 410
Lymphocytes, cell ×103/mm3 6.3 10.3 13.8 9.4 16.1 13.4 11.6
LDH, UI/L 377 441 254 208 410 – 417
D-dimer, ng/mL 796 2033 1287 1517 728 8000 NA
Fibrinogen, mg/dL 351 241 685 361 382 450
Ferritin, ng/mL 2167 3506 903 1010 1000 NA NA
PT, INR 1.06 1.13 0.94 1.33 1.13 1.18
aPTT, s 29.3 16.6 38.6 33.9 40.7 28.2
C-reactive protein, md/dL 0.62 1.52 1.97 12.23 2.28 13 4.73
Treatment during hospitalization
HCQ No Yes No No No No No
Lopinavir/ritonavir Yes Yes Yes No No No No
Remdesivir – – Yes – Yes Yes Yes
Steroids Methyl-prednisolone Methyl-prednisolone No No No Desametasone Desametasone
LMWH prophylaxis No Yes Yes Yes Yes Yes Yes
LMWH treatment 8000 UI ×2/day No No 6000 UI ×2/day 6000 UI ×2/day 6000 UI ×2/day 6000 UI ×2/day
Need for oxygen supplement Yes (cPAP) Yes (cPAP) Yes (VM with FiO2 40%) Yes (OTI) Yes (OTI) Yes (NIV) Yes (OTI)
IPH: ilio-psoas haematoma; LDH: lactic dehydrogenase; NA: not available; PT: prothrombin time; INR: international normalized ratio; aPTT: activated partial thromboplastin time; HCQ: hydroxychloroquine; LMWH: low molecular weight heparin; cPAP: continuous positive airway pressure; VM: venturi mask; OTI: orotracheal intubation; NIV: non-invasive ventilation.
Four patients were female with an age ranging between 65 and 80 years and a median body mass index of 28 (IQR 25–32). At the time of COVID-19 diagnosis, all patients were diagnosed with pneumonia and had at least one comorbidity; among them, three patients had hypertension, two patients had diabetes and one patient a chronic obstructive pulmonary disease (COPD). The median hospital stay was 34 days (IQR 30–36). Severe clinical presentation was observed in four patients with an admission oxygen saturation (SpO2) 93% on room air (IQR 90–98). The arterial oxygen partial pressure (PaO2 in mmHg) to fractional inspired oxygen (PaO2/FiO2) ratio was between 200 mmHg and 300 mmHg and it required supplemental oxygen therapy with non-invasive ventilation by using continuous positive airway pressure (cPAP). Three patients required invasive mechanical ventilation with orotracheal intubation (OTI) and admission to the ICU. Chest CT scan performed at admission, showed bilateral ground-glass opacities (GGOs) and sub-segmental consolidations, mostly located in the peripheral zone. Moreover, other imaging features such as linear opacities, “crazy-paving” pattern, the “reverse halo sign” and subsegmental vessel enlargement were described. In particular, the vessel enlargement was described close to the GGOs, which is compatible with thrombo-inflammatory processes. The contrast enhancement scan showed in five out of seven thrombosis-mediated micro-perfusion defects in peripheral pulmonary vessels.
All patients received prophylactic low weight molecular heparin (LWMH) at admission, with the exception of one patient who started anticoagulant therapy with LWMH for pulmonary embolism previously diagnosed at the emergency department. The heparin dosages were modified according to clinical worsening and to the diagnosis of micro or macro pulmonary thromboembolic events. Only one patient was taking antiaggregants medications discontinued as soon as the diagnosis of IPH was made.
Overall, steroid therapy was administered in five patients. Specifically, concerning SARS-CoV-2 therapy, four patients received oral lopinavir/ritonavir (LPV/r, 400/100 mg twice per day for 14 days), one of them received also oral hydroxychloroquine (200 mg twice per day for 10 days); four patients were treated with intravenous remdesivir (200 mg on day one followed by 100 mg since day 2 to day 10), combined with intravenous dexamethasone (6 mg once daily for 10 days) in three of them. Inflammation and coagulation parameters are shown in Table 1. Briefly, all patients had an hyperinflammation pattern with a median ferritin level of 548 pg/mL (IQR 369–2167 pg/mL) and C-reactive protein (CRP) of 8.6 mg/dL (IQR 6.6–15.8 mg/dL) and a D-dimer of 796 ng/mL (IQR 608–1373 ng/mL).
During the hospitalization, after a median of 35 (IQR 29–50) days, signs of neurologic compression with hypoesthesia of the lower limbs and general signs of hypovolaemia and anaemia were always present. A CT scan of the abdomen was performed revealing iliopsoas haematoma in all patients apart from one subject studied by performing an MRI.
The patients were treated in a conservative manner, resuscitated, according to clinical judgement, with intravenous fluid, transfusions of red blood cells and other supportive measures; only one patient, haemodynamically unstable, underwent arterial embolization.
Five patients were discharged with no long-lasting complications, one patient is still hospitalized and one patient died.
Imaging
The differential diagnosis of pelvic mass of the abdominal wall includes the most frequent pathologies as sarcoma, haematoma and abscess. Mass signal characteristics, vascular pattern during and after intravenous injection of contrast media and the presence of other elements (e.g. small intralesional calcifications) might help the radiologist to perform a correct diagnosis.
In our cases, a non-enhanced CT scan of the abdomen showed diffuse enlargement as well as heterogeneous density of the ilio-psoas muscle with an area of high density or fluid-fluid level for subacute haematoma. The absence of a contiguous mass with a vascular pattern led to rule out a diagnosis of haemorrhagic sarcoma.
The size of the haematomas found ranged from a minimum of 4–10 cm. In four patients, the dimensions exceeded 12 cm of transverse diameter and longitudinal extension of about 10 cm.
In two cases, the CT angiogram showed multiple streaks of contrast blush; delayed phase shows pooling (increased size of blush) of intravenous contrast media (Figure 1(a,b)). Contrast media blush, lack of internal fluid density and marginal enhancing component may exclude the diagnosis of iliopsoas abscess and raises the possibility of subacute haematoma.
Figure 1. (a) Arterial phase CT of right haematoma and (b) delayed phase CT of right haematoma.
In one case, MRI study was performed and showed a heterogeneous iliopsoas mass, with normal diffusion-weighted imaging (DWI) pattern and hyperintense areas on T1- and T2-weighted images for haemoglobin catabolites (metaHb) for subacute haematoma (Figure 2(a–c)).
Figure 2. (a) MRI of right iliopsoas haematoma, (b) MRI of right iliopsoas haematoma and (c) MRI of right iliopsoas haematoma.
The MRI findings correlated with the evolution of the haematoma:Acute haematoma: Iso-hypointense in T1-weighted (w) or slightly hypointense to muscle and hypo-hyperintense in T2 weighted (w).
Subacute haematoma: Hyperintense in T1w and T2w; high intensity rim, higher intensity peripheral zone and lower intensity core in T1w and relatively higher signal from core to periphery in T2w.
Chronic haematoma: Hypointense rim in T1w and T2w.
Discussion
Spontaneous IPH is defined as a retroperitoneal collection of blood involving the ilio-psoas muscle. Few studies evaluated the incidence of spontaneous IPHs in patients undergoing anticoagulation therapy that has been reported ranging from 0.1 to 0.6% [12]. Recently, a retrospective study in no COVID patients showed an incidence of IPHs of 3.8 cases per 1000 admission in ICUs [9] and our incidence of 7.6 cases over 1000 hospitalization, is higher than that previously reported. Risk factors in our small case series are the same reported in literature, as age, anticoagulation, a high body mass index, comorbidities such as hypertension and diabetes [9,13,14], moreover, our patients had an activated partial thromboplastin time (aPTT) above the therapeutic range at the moment of the diagnosis of the IPHs. Disseminated intravascular coagulopathy (DIC) and increased aPTT are both independent predictors of unfavourable prognosis [9]. The precise pathogenesis of retroperitoneal bleeding is unknown, it is most commonly reported as a complication of anticoagulation and, more rarely, in the setting of a clotting disorder or traumatic injury during the patients’ mobilization in the prone position [15–19].
Due to the anatomical proximity of these muscles to the lumbar plexus, it is hypothesized that retroperitoneal pre-existing microvascular atherosclerosis could increase sensitivity to rupture and microtrauma such as cough or vomiting could also lead to retroperitoneal bleeding [20].
It is well known that one of the causes of mortality in COVID-19 patients is venous thromboembolism (VTE) as evidenced by altered coagulation profile like elevated D-dimers [1,20].
This hypercoagulable phenomenon is due to the increased pro-inflammatory cytokines leading to atherosclerotic changes through local inflammation, microvascular thrombi and haemodynamic changes with multiorgan failure and death [1]. For these reasons, the administration of prophylactic or therapeutic anticoagulant agents is recommended [3–8]. Five cases among those we reported were fully treated with low molecular weight heparin for pulmonary micro-thrombosis.
Anticoagulation treatment and additional anti-platelet medications, increase the risk of major bleeding complications like retroperitoneal haemorrhage [9,10]. In a French ICU, 19 out of 92 (21%) COVID-19 patients on full anticoagulant treatment had 22 haemorrhagic events, and five of them were gastrointestinal [21].
Regarding IPHs, the optimal treatment remains controversial; however, initial treatment consists of discontinuation of anticoagulant agents, transfusion therapy, volume resuscitation and supportive measures; haemodynamically unstable patients, according to expert clinical judgement, can be treated with arterial embolization as it is minimally invasive with quick therapeutic effect when compared with surgical treatment.
Furthermore, stopping anticoagulation in case of IPHs and of a documented pulmonary micro-thrombosis in COVID-19 could be life-threatening.
Given the indications to prescribe anticoagulation in COVID-19 and the lack of solid evidences on the optimal dose and duration, specifically in micro-thrombosis, it is important to be aware of the iliopsoas haematoma as a potentially serious complication. As there is no consensus on therapeutical management of IPHs, each decision (i.e. conservative treatment, embolization, surgical or CT scan-guided haematoma’s drainage) should be made according to the clinical stability of the patients and by weighting risks and benefits.
We definitely need more studies in order to establish which is the optimal heparin dose and how long it should be continued in micro-thrombosis, as anticoagulation may increase the risk of major and potentially fatal bleeding.
Disclosure statement
All the authors have no competing interests that might be perceived to influence the results and/or discussion reported in this paper. Outside of this submitted work: Alessandra Vergori received institutional grant from Gilead Sciences, personal fees and travel grant from Janssen, personal fee from MSD; Andrea Antinori has served as a paid consultant to Gilead Sciences, Janssen-Cilag, Merck and ViiV Healthcare and received research institutional grants from Gilead Sciences, Janssen-Cilag and ViiV Healthcare. The other co-authors declare no conflicts of interests outside the submitted work.
Data availability statement
Data available on request from the authors.
Appendix The authors gratefully acknowledge nurse staff, all the patients and all members of the ReCOVeRI Study Group: Maria Alessandra Abbonizio, Amina Abdeddaim, Elisabetta Agostini, Chiara Agrati, Fabrizio Albarello, Gioia Amadei, Alessandra Amendola, Andrea Antinori, Maria Assunta Antonica, Mario Antonini, Tommaso Ascoli Bartoli, Francesco Baldini, Raffaella Barbaro, Barbara Bartolini, Rita Bellagamba, Martina Benigni, Nazario Bevilacqua, Gianluigi Biava, Michele Bibas, Licia Bordi, Veronica Bordoni, Evangelo Boumis, Marta Branca, Rosanna Buonomo, Donatella Busso, Marta Camici, Paolo Campioni, Flaminia Canichella, Maria Rosaria Capobianchi, Alessandro Capone, Cinzia Caporale, Emanuela Caraffa, Ilaria Caravella, Fabrizio Carletti, Concetta Castilletti, Adriana Cataldo, Stefano Cerilli, Carlotta Cerva, Roberta Chiappini, Pierangelo Chinello, Maria Assunta Cianfarani, Carmine Ciaralli, Claudia Cimaglia, Nicola Cinicola, Veronica Ciotti, Stefania Cicalini, Francesca Colavita, Angela Corpolongo, Massimo Cristofaro, Salvatore Curiale, Alessandra D’Abramo, Cristina Dantimi, Alessia De Angelis, Giada De Angelis, Maria Grazia De Palo, Federico De Zottis, Virginia Di Bari, Rachele Di Lorenzo, Federica Di Stefano, Gianpiero D’Offizi, Davide Donno, Francesca Evangelista, Francesca Faraglia, Anna Farina, Federica Ferraro, Lorena Fiorentini, Andrea Frustaci, Matteo Fusetti, Vincenzo Galati, Roberta Gagliardini, Paola Gallì, Gabriele Garotto, Ilaria Gaviano, Saba Gebremeskel Tekle, Maria Letizia Giancola, Filippo Giansante, Emanuela Giombini, Guido Granata, Maria Cristina Greci, Elisabetta Grilli, Susanna Grisetti, Gina Gualano, Fabio Iacomi, Marta Iaconi, Giuseppina Iannicelli, Carlo Inversi, Giuseppe Ippolito, Eleonora Lalle, Maria Elena Lamanna, Simone Lanini, Daniele Lapa, Luciana Lepore, Raffaella Libertone, Raffaella Lionetti, Giuseppina Liuzzi, Laura Loiacono, Andrea Lucia, Franco Lufrani, Manuela Macchione, Gaetano Maffongelli, Alessandra Marani, Luisa Marchioni, Andrea Mariano, Maria Cristina Marini, Micaela Maritti, Annelisa Mastrobattista, Ilaria Mastrorosa, Giulia Matusali, Valentina Mazzotta, Paola Mencarini, Silvia Meschi, Francesco Messina, Sibiana Micarelli, Giulia Mogavero, Annalisa Mondi, Marzia Montalbano, Chiara Montaldo, Silvia Mosti, Silvia Murachelli, Maria Musso, Michela Nardi, Assunta Navarra, Emanuele Nicastri, Martina Nocioni, Pasquale Noto, Roberto Noto, Alessandra Oliva, Ilaria Onnis, Sandrine Ottou, Claudia Palazzolo, Emanuele Pallini, Fabrizio Palmieri, Giulio Palombi, Carlo Pareo, Virgilio Passeri, Federico Pelliccioni, Giovanna Penna, Antonella Petrecchia, Ada Petrone, Nicola Petrosillo, Elisa Pianura, Carmela Pinnetti, Maria Pisciotta, Pierluca Piselli, Silvia Pittalis, Agostina Pontarelli, Costanza Proietti, Vincenzo Puro, Paolo Migliorisi Ramazzini, Alessia Rianda, Gabriele Rinonapoli, Silvia Rosati, Dorotea Rubino, Martina Rueca, Alberto Ruggeri, Alessandra Sacchi, Alessandro Sampaolesi, Francesco Sanasi, Carmen Santagata, Alessandra Scarabello, Silvana Scarcia, Vincenzo Schininà, Paola Scognamiglio, Laura Scorzolini, Giulia Stazi, Giacomo Strano,Fabrizio Taglietti, Chiara Taibi, Giorgia Taloni, Tetaj Nardi, Roberto Tonnarini, Simone Topino, Martina Tozzi, Francesco Vaia, Francesco Vairo, Maria Beatrice Valli, Alessandra Vergori, Laura Vincenzi, Ubaldo Visco-Comandini, Serena Vita, Pietro Vittozzi, Mauro Zaccarelli, Antonella Zanetti and Sara Zito. | Oral | DrugAdministrationRoute | CC BY | 33491498 | 19,026,066 | 2021-12 |
What was the dosage of drug 'LOPINAVIR\RITONAVIR'? | Spontaneous ilio-psoas haematomas (IPHs): a warning for COVID-19 inpatients.
Critically ill patients with COVID-19 are at increased risk of developing a hypercoagulable state due to haemostatic changes directly related to the SARS-CoV-2 infection or to the consequence of the cytokine storm. Anticoagulation is now recommended to reduce the thrombotic risk. Ilio-psoas haematoma (IPH) is a potentially lethal condition that can arise during the hospitalization, especially in intensive care units (ICUs) and frequently reported as a complication of anticoagulation treatment.
We report a case series of seven subjects with SARS-CoV-2 pneumonia complicated by Ilio-psoas haematomas (IPHs) at our COVID-Hospital in Rome, Italy.
Over the observation period, 925 subjects with confirmed SARS-CoV-2 infection were admitted to our COVID-hospital. Among them, we found seven spontaneous IPHs with an incidence of 7.6 cases per 1000 hospitalization. All the reported cases had a severe manifestation of COVID-19 pneumonia, with at least one comorbidity and 5/7 were on treatment with low weight molecular heparin for micro or macro pulmonary thrombosis.
Given the indications to prescribe anticoagulant therapy in COVID-19 and the lack of solid evidences on the optimal dose and duration, it is important to be aware of the iliopsoas haematoma as a potentially serious complication in COVID-19 inpatients. KEY MESSAGE Critically ill patients with COVID-19 are at increased risk of hypercoagulability state and anticoagulation therapy is recommended. Ilio-psoas haematoma (IPH) is found to be a complication of anticoagulation regimen especially in severe COVID-19 cases. An incidence of 7.6 cases per 1000 admission of IPHs was reported. Hypoesthesia of the lower limbs, pain triggered by femoral rotation, hypovolaemia and anaemia are the most common symptoms and signs of IPHs that should alert physician.
Introduction
Critically ill patients with coronavirus disease-2019 (COVID-19) are at increased risk of developing a hypercoagulable state [1].
The pathophysiology behind this phenomenon has been suspected to be a result of haemostatic changes that might be direct effect of severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2) or a consequence of a cytokine storm that alters the onset of the systemic inflammatory response syndrome (SIRS), as observed in other viral disease [2]. To hinder this hypercoagulable state, COVID-19 patients should be properly anticoagulated to reduce the thrombotic risk [3–8].
IPH is a potentially lethal condition frequently reported as a complication of anticoagulation therapy that can arise during the hospitalization, especially in intensive care units (ICUs) [9]. To date, a single case of spontaneous IPH in a COVID-19 patient has been reported [10]. Here, we report seven cases of spontaneous IPHs occurring in patients with severe COVID-19 pneumonia admitted to our Institute.
Materials and methods
Study population and setting
We consecutively included all subjects with a microbiologically confirmed SARS-CoV-2 infection who were admitted to the National Institute for Infectious Diseases IRCCS Lazzaro Spallanzani in Rome, Italy, between 1 March 2020 and 30 October 2020. Medical history, demographic and clinical data were collected through review of medical records. Data have been collected for the ReCOVeRI Study, a registry on COVID-19 for clinical Research of the National Institute for Infectious Diseases L. Spallanzani, approved by the Ethical Committee of the National Institute for Infectious Diseases L. Spallanzani IRCCS (number 164, 26 June 2020). Laboratory and radiologic assessments during the hospital stay were performed by the treating physician according to the hospital operative procedure. All patients gave informed consent for collecting personal data for research purposes.
Definitions
A confirmed case of COVID-19 was defined by a positive real-time reverse-transcription PCR (RT-PCR) assay for SARS-CoV-2 on nasopharyngeal swab and/or a positive serology for SARS-CoV-2 (positive immunoglobulin (Ig) G or M or A for SARS-CoV-2).
Severe disease was defined as clinical signs of pneumonia plus one of the following: respiratory rate greater than 30 breaths per min, severe respiratory distress or oxygen saturation less than 90% on room air [11].
Hyperinflammation syndrome was defined as having at least two among D-dimer above 1000 ng/mL, ferritin above 500 mcg/L, LDH above 300 UI/L and lymphocyte count below 1000 cell/mm3 [1]. Ilio-psoas haematoma was diagnosed by using computerized tomography (CT)-scan or magnetic resonance imaging (MRI).
All CT scans were performed on a multi-detector row CT scanner (Bright Speed, General Electric Medical Systems, Milwaukee, WI) using 120 kV pp, 250 mA, pitch of 1.375, gantry rotation time of 0.6 s. A chest and abdomen CT was performed from the apex of lung to symphysis pubis before and after injection of iodinated contrast media into a peripheral vein with three-phase arterial, venous and delayed phase. The baseline scan of the thorax was reconstructed with slice thicknesses of 0.625 mm and spacing of 1 mm with high contrast resolution algorithm. The contrast media scan of the thorax and abdomen was reconstructed with slice thicknesses of 1.25 mm and spacing of 1 mm, completed with multiplanar reconstructions (MPR and Mip). The MRI scan was performed on HDx scanner (General Electric Medical Systems, Milwaukee, WI), 1.5 T, with T1 and T2 weighted sequences also after spectral fat subtraction and paramagnetic contrast media.
Results
Over the observation period, 925 subjects with confirmed SARS-CoV-2 infection were admitted to our COVID-hospital. Among them, we observed seven spontaneous IPHs with an incidence of 7.6 cases per 1000 hospitalization.
Clinical characteristics and outcome
Main demographic and clinical characteristics and outcomes are summarized in Table 1.
Table 1. Characteristics of the reported cases.
Patient 1 Patient 2 Patient 3 Patient 4 Patient 5 Patient 6 Patient 7
Age 80 75 94 55 64 72 66
Gender M F F M F M F
Comorbidities Atrial fibrillation; COPD Hypertension; asthma Diabetes, ischaemic heart disease, hypertension, cognitive impairment, bedridden syndrome Not reported Ischaemic heart disease, obesity Diabetes, hypertension Thyroid disease
Clinical presentation at admission
Days from symptoms onset to admission 14 5 2 14 7 7 7
Symptoms Arthralgia, asthenia, dyspnoea, fever, cough Headache, dyspnoea, fever, cough, myalgia Fever, cough Fever, cough Fever, cough Fever, cough, dyspnoea Fever, dysgeusia, cough
SpO2 at admission on room air 90% 96% 95% 93% 91% 90% 92%
Days from admission to IPH 9 25 14 5 38 30 18
Days from admission to prophylactic anticoagulation start 1 9 0 0 0 0 0
Length of hospital stay (days) 20 50 36 34 – 70 29
Exitus Died Discharged Discharged Discharged Still admitted Discharged Discharged
Inflammatory Index at admission
Haemoglobin, g/dL 13.9 11.6 12.8 7.4 11 12.3 13.8
Platelets, ×103/mm3 121 142 185 176 255 193 350
Lymphocytes, cell ×103/mm3 4.2 9.6 12.2 15.0 8.1 7.5 12.21
LDH, UI/L 404 576 254 249 325 205 433
D-dimer, ng/mL 796 740 551 1373 608 10,850 901
Fibrinogen, mg/dL 798 487 47 314 824 531 706
ferritin, ng/mL 2167 4754 1113 548 303 369 534
PT, INR 0.93 1.08 1.23 1.18 1.06 1.12 1.21
PTT, sec 48.4 35.7 48.7 33.1 24.6 26.3 26.6
C-reactive protein, md/dL 15.78 8.62 33.1 16.2 12.9 4.5 6.6
Inflammatory index at ilio-psoas haematoma
Haemoglobin, g/dL 8.7 6.8 10.5 8 6.9 4.9 8.8
Platelet, ×103/mm3 234 91 295 173 492 217 410
Lymphocytes, cell ×103/mm3 6.3 10.3 13.8 9.4 16.1 13.4 11.6
LDH, UI/L 377 441 254 208 410 – 417
D-dimer, ng/mL 796 2033 1287 1517 728 8000 NA
Fibrinogen, mg/dL 351 241 685 361 382 450
Ferritin, ng/mL 2167 3506 903 1010 1000 NA NA
PT, INR 1.06 1.13 0.94 1.33 1.13 1.18
aPTT, s 29.3 16.6 38.6 33.9 40.7 28.2
C-reactive protein, md/dL 0.62 1.52 1.97 12.23 2.28 13 4.73
Treatment during hospitalization
HCQ No Yes No No No No No
Lopinavir/ritonavir Yes Yes Yes No No No No
Remdesivir – – Yes – Yes Yes Yes
Steroids Methyl-prednisolone Methyl-prednisolone No No No Desametasone Desametasone
LMWH prophylaxis No Yes Yes Yes Yes Yes Yes
LMWH treatment 8000 UI ×2/day No No 6000 UI ×2/day 6000 UI ×2/day 6000 UI ×2/day 6000 UI ×2/day
Need for oxygen supplement Yes (cPAP) Yes (cPAP) Yes (VM with FiO2 40%) Yes (OTI) Yes (OTI) Yes (NIV) Yes (OTI)
IPH: ilio-psoas haematoma; LDH: lactic dehydrogenase; NA: not available; PT: prothrombin time; INR: international normalized ratio; aPTT: activated partial thromboplastin time; HCQ: hydroxychloroquine; LMWH: low molecular weight heparin; cPAP: continuous positive airway pressure; VM: venturi mask; OTI: orotracheal intubation; NIV: non-invasive ventilation.
Four patients were female with an age ranging between 65 and 80 years and a median body mass index of 28 (IQR 25–32). At the time of COVID-19 diagnosis, all patients were diagnosed with pneumonia and had at least one comorbidity; among them, three patients had hypertension, two patients had diabetes and one patient a chronic obstructive pulmonary disease (COPD). The median hospital stay was 34 days (IQR 30–36). Severe clinical presentation was observed in four patients with an admission oxygen saturation (SpO2) 93% on room air (IQR 90–98). The arterial oxygen partial pressure (PaO2 in mmHg) to fractional inspired oxygen (PaO2/FiO2) ratio was between 200 mmHg and 300 mmHg and it required supplemental oxygen therapy with non-invasive ventilation by using continuous positive airway pressure (cPAP). Three patients required invasive mechanical ventilation with orotracheal intubation (OTI) and admission to the ICU. Chest CT scan performed at admission, showed bilateral ground-glass opacities (GGOs) and sub-segmental consolidations, mostly located in the peripheral zone. Moreover, other imaging features such as linear opacities, “crazy-paving” pattern, the “reverse halo sign” and subsegmental vessel enlargement were described. In particular, the vessel enlargement was described close to the GGOs, which is compatible with thrombo-inflammatory processes. The contrast enhancement scan showed in five out of seven thrombosis-mediated micro-perfusion defects in peripheral pulmonary vessels.
All patients received prophylactic low weight molecular heparin (LWMH) at admission, with the exception of one patient who started anticoagulant therapy with LWMH for pulmonary embolism previously diagnosed at the emergency department. The heparin dosages were modified according to clinical worsening and to the diagnosis of micro or macro pulmonary thromboembolic events. Only one patient was taking antiaggregants medications discontinued as soon as the diagnosis of IPH was made.
Overall, steroid therapy was administered in five patients. Specifically, concerning SARS-CoV-2 therapy, four patients received oral lopinavir/ritonavir (LPV/r, 400/100 mg twice per day for 14 days), one of them received also oral hydroxychloroquine (200 mg twice per day for 10 days); four patients were treated with intravenous remdesivir (200 mg on day one followed by 100 mg since day 2 to day 10), combined with intravenous dexamethasone (6 mg once daily for 10 days) in three of them. Inflammation and coagulation parameters are shown in Table 1. Briefly, all patients had an hyperinflammation pattern with a median ferritin level of 548 pg/mL (IQR 369–2167 pg/mL) and C-reactive protein (CRP) of 8.6 mg/dL (IQR 6.6–15.8 mg/dL) and a D-dimer of 796 ng/mL (IQR 608–1373 ng/mL).
During the hospitalization, after a median of 35 (IQR 29–50) days, signs of neurologic compression with hypoesthesia of the lower limbs and general signs of hypovolaemia and anaemia were always present. A CT scan of the abdomen was performed revealing iliopsoas haematoma in all patients apart from one subject studied by performing an MRI.
The patients were treated in a conservative manner, resuscitated, according to clinical judgement, with intravenous fluid, transfusions of red blood cells and other supportive measures; only one patient, haemodynamically unstable, underwent arterial embolization.
Five patients were discharged with no long-lasting complications, one patient is still hospitalized and one patient died.
Imaging
The differential diagnosis of pelvic mass of the abdominal wall includes the most frequent pathologies as sarcoma, haematoma and abscess. Mass signal characteristics, vascular pattern during and after intravenous injection of contrast media and the presence of other elements (e.g. small intralesional calcifications) might help the radiologist to perform a correct diagnosis.
In our cases, a non-enhanced CT scan of the abdomen showed diffuse enlargement as well as heterogeneous density of the ilio-psoas muscle with an area of high density or fluid-fluid level for subacute haematoma. The absence of a contiguous mass with a vascular pattern led to rule out a diagnosis of haemorrhagic sarcoma.
The size of the haematomas found ranged from a minimum of 4–10 cm. In four patients, the dimensions exceeded 12 cm of transverse diameter and longitudinal extension of about 10 cm.
In two cases, the CT angiogram showed multiple streaks of contrast blush; delayed phase shows pooling (increased size of blush) of intravenous contrast media (Figure 1(a,b)). Contrast media blush, lack of internal fluid density and marginal enhancing component may exclude the diagnosis of iliopsoas abscess and raises the possibility of subacute haematoma.
Figure 1. (a) Arterial phase CT of right haematoma and (b) delayed phase CT of right haematoma.
In one case, MRI study was performed and showed a heterogeneous iliopsoas mass, with normal diffusion-weighted imaging (DWI) pattern and hyperintense areas on T1- and T2-weighted images for haemoglobin catabolites (metaHb) for subacute haematoma (Figure 2(a–c)).
Figure 2. (a) MRI of right iliopsoas haematoma, (b) MRI of right iliopsoas haematoma and (c) MRI of right iliopsoas haematoma.
The MRI findings correlated with the evolution of the haematoma:Acute haematoma: Iso-hypointense in T1-weighted (w) or slightly hypointense to muscle and hypo-hyperintense in T2 weighted (w).
Subacute haematoma: Hyperintense in T1w and T2w; high intensity rim, higher intensity peripheral zone and lower intensity core in T1w and relatively higher signal from core to periphery in T2w.
Chronic haematoma: Hypointense rim in T1w and T2w.
Discussion
Spontaneous IPH is defined as a retroperitoneal collection of blood involving the ilio-psoas muscle. Few studies evaluated the incidence of spontaneous IPHs in patients undergoing anticoagulation therapy that has been reported ranging from 0.1 to 0.6% [12]. Recently, a retrospective study in no COVID patients showed an incidence of IPHs of 3.8 cases per 1000 admission in ICUs [9] and our incidence of 7.6 cases over 1000 hospitalization, is higher than that previously reported. Risk factors in our small case series are the same reported in literature, as age, anticoagulation, a high body mass index, comorbidities such as hypertension and diabetes [9,13,14], moreover, our patients had an activated partial thromboplastin time (aPTT) above the therapeutic range at the moment of the diagnosis of the IPHs. Disseminated intravascular coagulopathy (DIC) and increased aPTT are both independent predictors of unfavourable prognosis [9]. The precise pathogenesis of retroperitoneal bleeding is unknown, it is most commonly reported as a complication of anticoagulation and, more rarely, in the setting of a clotting disorder or traumatic injury during the patients’ mobilization in the prone position [15–19].
Due to the anatomical proximity of these muscles to the lumbar plexus, it is hypothesized that retroperitoneal pre-existing microvascular atherosclerosis could increase sensitivity to rupture and microtrauma such as cough or vomiting could also lead to retroperitoneal bleeding [20].
It is well known that one of the causes of mortality in COVID-19 patients is venous thromboembolism (VTE) as evidenced by altered coagulation profile like elevated D-dimers [1,20].
This hypercoagulable phenomenon is due to the increased pro-inflammatory cytokines leading to atherosclerotic changes through local inflammation, microvascular thrombi and haemodynamic changes with multiorgan failure and death [1]. For these reasons, the administration of prophylactic or therapeutic anticoagulant agents is recommended [3–8]. Five cases among those we reported were fully treated with low molecular weight heparin for pulmonary micro-thrombosis.
Anticoagulation treatment and additional anti-platelet medications, increase the risk of major bleeding complications like retroperitoneal haemorrhage [9,10]. In a French ICU, 19 out of 92 (21%) COVID-19 patients on full anticoagulant treatment had 22 haemorrhagic events, and five of them were gastrointestinal [21].
Regarding IPHs, the optimal treatment remains controversial; however, initial treatment consists of discontinuation of anticoagulant agents, transfusion therapy, volume resuscitation and supportive measures; haemodynamically unstable patients, according to expert clinical judgement, can be treated with arterial embolization as it is minimally invasive with quick therapeutic effect when compared with surgical treatment.
Furthermore, stopping anticoagulation in case of IPHs and of a documented pulmonary micro-thrombosis in COVID-19 could be life-threatening.
Given the indications to prescribe anticoagulation in COVID-19 and the lack of solid evidences on the optimal dose and duration, specifically in micro-thrombosis, it is important to be aware of the iliopsoas haematoma as a potentially serious complication. As there is no consensus on therapeutical management of IPHs, each decision (i.e. conservative treatment, embolization, surgical or CT scan-guided haematoma’s drainage) should be made according to the clinical stability of the patients and by weighting risks and benefits.
We definitely need more studies in order to establish which is the optimal heparin dose and how long it should be continued in micro-thrombosis, as anticoagulation may increase the risk of major and potentially fatal bleeding.
Disclosure statement
All the authors have no competing interests that might be perceived to influence the results and/or discussion reported in this paper. Outside of this submitted work: Alessandra Vergori received institutional grant from Gilead Sciences, personal fees and travel grant from Janssen, personal fee from MSD; Andrea Antinori has served as a paid consultant to Gilead Sciences, Janssen-Cilag, Merck and ViiV Healthcare and received research institutional grants from Gilead Sciences, Janssen-Cilag and ViiV Healthcare. The other co-authors declare no conflicts of interests outside the submitted work.
Data availability statement
Data available on request from the authors.
Appendix The authors gratefully acknowledge nurse staff, all the patients and all members of the ReCOVeRI Study Group: Maria Alessandra Abbonizio, Amina Abdeddaim, Elisabetta Agostini, Chiara Agrati, Fabrizio Albarello, Gioia Amadei, Alessandra Amendola, Andrea Antinori, Maria Assunta Antonica, Mario Antonini, Tommaso Ascoli Bartoli, Francesco Baldini, Raffaella Barbaro, Barbara Bartolini, Rita Bellagamba, Martina Benigni, Nazario Bevilacqua, Gianluigi Biava, Michele Bibas, Licia Bordi, Veronica Bordoni, Evangelo Boumis, Marta Branca, Rosanna Buonomo, Donatella Busso, Marta Camici, Paolo Campioni, Flaminia Canichella, Maria Rosaria Capobianchi, Alessandro Capone, Cinzia Caporale, Emanuela Caraffa, Ilaria Caravella, Fabrizio Carletti, Concetta Castilletti, Adriana Cataldo, Stefano Cerilli, Carlotta Cerva, Roberta Chiappini, Pierangelo Chinello, Maria Assunta Cianfarani, Carmine Ciaralli, Claudia Cimaglia, Nicola Cinicola, Veronica Ciotti, Stefania Cicalini, Francesca Colavita, Angela Corpolongo, Massimo Cristofaro, Salvatore Curiale, Alessandra D’Abramo, Cristina Dantimi, Alessia De Angelis, Giada De Angelis, Maria Grazia De Palo, Federico De Zottis, Virginia Di Bari, Rachele Di Lorenzo, Federica Di Stefano, Gianpiero D’Offizi, Davide Donno, Francesca Evangelista, Francesca Faraglia, Anna Farina, Federica Ferraro, Lorena Fiorentini, Andrea Frustaci, Matteo Fusetti, Vincenzo Galati, Roberta Gagliardini, Paola Gallì, Gabriele Garotto, Ilaria Gaviano, Saba Gebremeskel Tekle, Maria Letizia Giancola, Filippo Giansante, Emanuela Giombini, Guido Granata, Maria Cristina Greci, Elisabetta Grilli, Susanna Grisetti, Gina Gualano, Fabio Iacomi, Marta Iaconi, Giuseppina Iannicelli, Carlo Inversi, Giuseppe Ippolito, Eleonora Lalle, Maria Elena Lamanna, Simone Lanini, Daniele Lapa, Luciana Lepore, Raffaella Libertone, Raffaella Lionetti, Giuseppina Liuzzi, Laura Loiacono, Andrea Lucia, Franco Lufrani, Manuela Macchione, Gaetano Maffongelli, Alessandra Marani, Luisa Marchioni, Andrea Mariano, Maria Cristina Marini, Micaela Maritti, Annelisa Mastrobattista, Ilaria Mastrorosa, Giulia Matusali, Valentina Mazzotta, Paola Mencarini, Silvia Meschi, Francesco Messina, Sibiana Micarelli, Giulia Mogavero, Annalisa Mondi, Marzia Montalbano, Chiara Montaldo, Silvia Mosti, Silvia Murachelli, Maria Musso, Michela Nardi, Assunta Navarra, Emanuele Nicastri, Martina Nocioni, Pasquale Noto, Roberto Noto, Alessandra Oliva, Ilaria Onnis, Sandrine Ottou, Claudia Palazzolo, Emanuele Pallini, Fabrizio Palmieri, Giulio Palombi, Carlo Pareo, Virgilio Passeri, Federico Pelliccioni, Giovanna Penna, Antonella Petrecchia, Ada Petrone, Nicola Petrosillo, Elisa Pianura, Carmela Pinnetti, Maria Pisciotta, Pierluca Piselli, Silvia Pittalis, Agostina Pontarelli, Costanza Proietti, Vincenzo Puro, Paolo Migliorisi Ramazzini, Alessia Rianda, Gabriele Rinonapoli, Silvia Rosati, Dorotea Rubino, Martina Rueca, Alberto Ruggeri, Alessandra Sacchi, Alessandro Sampaolesi, Francesco Sanasi, Carmen Santagata, Alessandra Scarabello, Silvana Scarcia, Vincenzo Schininà, Paola Scognamiglio, Laura Scorzolini, Giulia Stazi, Giacomo Strano,Fabrizio Taglietti, Chiara Taibi, Giorgia Taloni, Tetaj Nardi, Roberto Tonnarini, Simone Topino, Martina Tozzi, Francesco Vaia, Francesco Vairo, Maria Beatrice Valli, Alessandra Vergori, Laura Vincenzi, Ubaldo Visco-Comandini, Serena Vita, Pietro Vittozzi, Mauro Zaccarelli, Antonella Zanetti and Sara Zito. | UNK (400/100MG TWICE PER DAY) | DrugDosageText | CC BY | 33491498 | 19,026,066 | 2021-12 |
What was the administration route of drug 'APIXABAN'? | A 78-Year-Old Man with a Pulmonary Embolism Who Developed Skin Necrosis 7 Days After Treatment with the Direct Oral Anticoagulant Factor Xa Inhibitor Apixaban.
BACKGROUND Apixaban is one of the newer direct oral anticoagulants (DOACs) being used to manage venous thrombosis. Skin toxicities are recognized adverse effects of the new DOACs, but are rare and usually associated with vasculitis. This report is of a 78-year-old man admitted to the hospital with pulmonary thromboembolism, who developed severe and extensive skin necrosis of both forearms 7 days after treatment with apixaban. CASE REPORT A 78-year-old man was admitted for pulmonary embolism and congestive heart failure exacerbation. He was started on therapeutic enoxaparin and diuresis. Later on, enoxaparin was substituted with apixaban. Seven days after starting apixaban, he suddenly developed skin changes that developed into skin necrosis on both forearms and the abdominal wall. A skin biopsy was not performed due to the high risk of bleeding. Skin necrosis was diagnosed based on clinical findings. A review of clinical data and the patient's medication profile did not reveal any other possible etiology or culprit medication. Clinical presentation and lab values were not consistent with infections or autoimmune etiologies. Apixaban was discontinued as it was perceived to be the likely cause of skin necrosis. The skin changes gradually improved within 1 week with supportive wound care, and the patient did not require a skin graft. The patient was discharged safely with subcutaneous low-molecular-weight heparin therapy. CONCLUSIONS This report shows that skin toxicity can be associated with apixaban and that with the increasing use of these newer DOACs, clinicians should be aware of these potential adverse effects.
Background
In late 2010, a new class of anticoagulants known as direct oral anticoagulants (DOAC) was introduced as an alternative to conventional anticoagulation therapy [1]. DOACs have proven to be equal or superior to conventional therapy for preventing stroke in nonvalvular atrial fibrillation, with similar efficacy for prevention and treatment of venous thromboembolism [1,2]. This new class has clear advantages over warfarin, with better pharmacodynamics and pharmacokinetic properties, less drug-drug interactions, and no requirement for frequent coagulation test monitoring [2,3]. Bleeding risk is a significant concern for all anticoagulant therapies, but trials showed that compared to warfarin, DOACs have similar or less risk of major bleeding [4,5]. DOACs include direct thrombin inhibitors (Dabigatran) and direct factor Xa inhibitors (rivaroxaban, apixaban, and edoxaban).
There are reported data available for non-hemorrhagic adverse events caused by DOACs, such as leukocytoclastic vasculitis, psoriatic-like skin reaction, and lichenoid reaction [6–9]. Although skin toxicities are recognized adverse effects of the new DOACs, including apixaban, they are rare and are usually associated with vasculitis. This report is of a 78-year-old man admitted to the hospital with pulmonary thromboembolism, who developed severe and extensive skin necrosis of both forearms 7 days after treatment with apixaban.
Case Report
A 78-year-old man with a history of hypertension, chronic obstructive pulmonary disease (COPD), coronary artery disease (CAD), coronary artery bypass graft (CABG), diastolic heart failure, and stage 3 chronic kidney disease (CKD) presented with worsening dyspnea, swelling of the left leg, and pain for the last 2 days. On physical examination, he had bilateral basilar crackles with expiratory wheezing and bilateral lower-extremity edema. Troponin was 0.02 ng/ml, pro brain natriuretic peptide (ProBNP) was 1140 pg/ml, and D-dimer was elevated to 1066 ng/ml FEU. A chest X-ray showed right basilar lung infiltrate. Initially, he was started on bronchodilators for his COPD, diuresis for his congestive heart failure (CHF), and intravenous antibiotics for possible pneumonia. Pulmonary embolism was also considered in the differential diagnosis due to his increased oxygen requirement. CT angiography of the chest was not performed due to his elevated creatinine. A ventilation-perfusion lung scan (V/Q scan) showed an intermediate probability of pulmonary embolism. Ultrasound Doppler imaging of the lower extremities did not show deep vein thrombosis. The combination of the patient’s acute clinical presentation, elevated D-dimer, and V/Q scan intermediate probability was suggestive of pulmonary embolism. Therapeutic subcutaneous enoxaparin was initiated based on the creatinine clearance, as recommended by the pharmacy.
During the hospital stay, his condition was stabilized with bronchodilators and systemic glucocorticoids for advanced COPD, piperacillin/tazobactam for pneumonia, and diuresis for chronic diastolic congestive heart failure. His respiration was supported with a continued nasal cannula and BiPAP as needed. After 2 days of enoxaparin, he was switched to oral apixaban for his pulmonary embolism. His shortness of breath was improving slowly due to his underlying advanced COPD and CHF. During this time, he was waiting for a skilled nursing home placement.
After the seventh day of initiating apixaban, he developed a painful small ecchymosis on the left forearm, which quickly progressed to skin necrosis across the left forearm within 2 days (see Figure 1). He also developed skin necrosis on the right arm and the abdominal wall area. Surgery and Dermatology services were consulted. A skin biopsy was not performed due to the high risk of bleeding. The clinical presentation was not consistent with any autoimmune process. Apixaban was discontinued immediately. After reviewing all the patient’s medications during that interval, the clinical conclusion was a diagnosis of apixaban-induced skin necrosis. The skin changes gradually improved within 7 days with wound care and dressing. The patient did not require skin grafting, and he was discharged with therapeutic subcutaneous enoxaparin. Two weeks after discharge, he was readmitted for anemia, at which time his skin lesion had improved significantly.
Discussion
Apixaban is reported to cause non-hemorrhagic adverse effects [6–9]. In leukocytoclastic vasculitis cases, lower-extremity skin purpura developed after 9 days of apixaban therapy [6,7]. It was managed with systemic steroid with complete resolution of skin changes within a few weeks [6,7]. In one case, the patient was successfully switched to rivaroxaban without any adverse events [7]. Veliyev et al. described a case of palmoplantar psoriasiform drug eruption in which cutaneous drug eruption developed after 3 days of apixaban therapy; it was managed with topical steroids, and skin changes resolved entirely within a few weeks [8]. Patil et al. reported a case of apixaban-induced delayed drug reaction-lichenoid reaction in which the patient was managed with topical steroids [9]. Similar to apixaban, rivaroxaban, another DOAC, has also been reported to cause skin toxicity. Soliman et al. reported a case of rivaroxaban-induced skin necrosis in which skin changes developed after 3 days of therapy; it was managed with a topical steroid, and the patient was successfully switched to low-molecular-weight heparin [10].
The pathophysiology behind DOAC-induced skin toxicity is unclear. Possible mechanisms include microvascular thrombosis, an imbalance between anticoagulant and procoagulant factors, autoimmune response, and direct medication toxicity [10]. The pathophysiology of warfarin-induced skin necrosis is well explained. Warfarin inactivates vitamin K-dependent coagulation factors II, VII, IX, and X. However, it also inactivates the vitamin K-dependent anticoagulants protein C and S. This may generate an imbalance between pro-coagulation and anticoagulation, which can lead to microvascular thrombi and skin necrosis [11,12]. It is unknown at this time if warfarin-induced skin necrosis and DOAC-induced skin toxicity share a similar pathophysiology.
There are some limitations to this case report. An autoimmune workup for drug-induced vasculitis was not performed. Protein C and Protein S levels could not be obtained because of inpatient restrictions as per hospital policy. A skin biopsy was not performed as the patient was on aspirin and Plavix as a home medication due to his significant coronary artery disease, and he was at high risk for bleeding and post-biopsy skin infections.
Our experience and literature review suggest that treatment of this condition with prompt discontinuation of the offending DOAC and conservative management with wound care is sufficient to manage the skin necrosis.
Conclusions
Although skin toxicities are recognized adverse effects of the new DOACs, including apixaban, they are rare and are usually associated with vasculitis. To the best of our knowledge, apixaban-induced skin necrosis has not been reported. This report has shown that skin toxicity can be associated with apixaban and that with the increasing use of these newer DOACs, clinicians should be aware of these potential adverse effects. In this case, the mainstay of treatment was the discontinuation of the causative agent and supportive treatment.
Disclaimer
This whole case report was supported by HCA Health Care-Kingwood Medical Center. The views expressed in this publication represent those of the authors and do not necessarily represent the official views of HCA Healthcare or any of its affiliated entities.
Conflict of interest
None
Abbreviations
DOACsdirect oral anticoagulants;
COPDchronic obstructive pulmonary disease;
CADcoronary artery disease;
CABGcoronary artery bypass graft;
CKDchronic kidney disease;
ProBNPpro brain natriuretic peptide;
V/Q scanventilation-perfusion scan;
AISNapixaban-induced skin necrosis
Figure 1. The left forearm shows a confluent and extensive area of skin necrosis with no raised skin lesions, no vesicles, and no purulent exudate. | Oral | DrugAdministrationRoute | CC BY-NC-ND | 33493142 | 18,897,969 | 2021-01-25 |
What was the administration route of drug 'ENOXAPARIN'? | A 78-Year-Old Man with a Pulmonary Embolism Who Developed Skin Necrosis 7 Days After Treatment with the Direct Oral Anticoagulant Factor Xa Inhibitor Apixaban.
BACKGROUND Apixaban is one of the newer direct oral anticoagulants (DOACs) being used to manage venous thrombosis. Skin toxicities are recognized adverse effects of the new DOACs, but are rare and usually associated with vasculitis. This report is of a 78-year-old man admitted to the hospital with pulmonary thromboembolism, who developed severe and extensive skin necrosis of both forearms 7 days after treatment with apixaban. CASE REPORT A 78-year-old man was admitted for pulmonary embolism and congestive heart failure exacerbation. He was started on therapeutic enoxaparin and diuresis. Later on, enoxaparin was substituted with apixaban. Seven days after starting apixaban, he suddenly developed skin changes that developed into skin necrosis on both forearms and the abdominal wall. A skin biopsy was not performed due to the high risk of bleeding. Skin necrosis was diagnosed based on clinical findings. A review of clinical data and the patient's medication profile did not reveal any other possible etiology or culprit medication. Clinical presentation and lab values were not consistent with infections or autoimmune etiologies. Apixaban was discontinued as it was perceived to be the likely cause of skin necrosis. The skin changes gradually improved within 1 week with supportive wound care, and the patient did not require a skin graft. The patient was discharged safely with subcutaneous low-molecular-weight heparin therapy. CONCLUSIONS This report shows that skin toxicity can be associated with apixaban and that with the increasing use of these newer DOACs, clinicians should be aware of these potential adverse effects.
Background
In late 2010, a new class of anticoagulants known as direct oral anticoagulants (DOAC) was introduced as an alternative to conventional anticoagulation therapy [1]. DOACs have proven to be equal or superior to conventional therapy for preventing stroke in nonvalvular atrial fibrillation, with similar efficacy for prevention and treatment of venous thromboembolism [1,2]. This new class has clear advantages over warfarin, with better pharmacodynamics and pharmacokinetic properties, less drug-drug interactions, and no requirement for frequent coagulation test monitoring [2,3]. Bleeding risk is a significant concern for all anticoagulant therapies, but trials showed that compared to warfarin, DOACs have similar or less risk of major bleeding [4,5]. DOACs include direct thrombin inhibitors (Dabigatran) and direct factor Xa inhibitors (rivaroxaban, apixaban, and edoxaban).
There are reported data available for non-hemorrhagic adverse events caused by DOACs, such as leukocytoclastic vasculitis, psoriatic-like skin reaction, and lichenoid reaction [6–9]. Although skin toxicities are recognized adverse effects of the new DOACs, including apixaban, they are rare and are usually associated with vasculitis. This report is of a 78-year-old man admitted to the hospital with pulmonary thromboembolism, who developed severe and extensive skin necrosis of both forearms 7 days after treatment with apixaban.
Case Report
A 78-year-old man with a history of hypertension, chronic obstructive pulmonary disease (COPD), coronary artery disease (CAD), coronary artery bypass graft (CABG), diastolic heart failure, and stage 3 chronic kidney disease (CKD) presented with worsening dyspnea, swelling of the left leg, and pain for the last 2 days. On physical examination, he had bilateral basilar crackles with expiratory wheezing and bilateral lower-extremity edema. Troponin was 0.02 ng/ml, pro brain natriuretic peptide (ProBNP) was 1140 pg/ml, and D-dimer was elevated to 1066 ng/ml FEU. A chest X-ray showed right basilar lung infiltrate. Initially, he was started on bronchodilators for his COPD, diuresis for his congestive heart failure (CHF), and intravenous antibiotics for possible pneumonia. Pulmonary embolism was also considered in the differential diagnosis due to his increased oxygen requirement. CT angiography of the chest was not performed due to his elevated creatinine. A ventilation-perfusion lung scan (V/Q scan) showed an intermediate probability of pulmonary embolism. Ultrasound Doppler imaging of the lower extremities did not show deep vein thrombosis. The combination of the patient’s acute clinical presentation, elevated D-dimer, and V/Q scan intermediate probability was suggestive of pulmonary embolism. Therapeutic subcutaneous enoxaparin was initiated based on the creatinine clearance, as recommended by the pharmacy.
During the hospital stay, his condition was stabilized with bronchodilators and systemic glucocorticoids for advanced COPD, piperacillin/tazobactam for pneumonia, and diuresis for chronic diastolic congestive heart failure. His respiration was supported with a continued nasal cannula and BiPAP as needed. After 2 days of enoxaparin, he was switched to oral apixaban for his pulmonary embolism. His shortness of breath was improving slowly due to his underlying advanced COPD and CHF. During this time, he was waiting for a skilled nursing home placement.
After the seventh day of initiating apixaban, he developed a painful small ecchymosis on the left forearm, which quickly progressed to skin necrosis across the left forearm within 2 days (see Figure 1). He also developed skin necrosis on the right arm and the abdominal wall area. Surgery and Dermatology services were consulted. A skin biopsy was not performed due to the high risk of bleeding. The clinical presentation was not consistent with any autoimmune process. Apixaban was discontinued immediately. After reviewing all the patient’s medications during that interval, the clinical conclusion was a diagnosis of apixaban-induced skin necrosis. The skin changes gradually improved within 7 days with wound care and dressing. The patient did not require skin grafting, and he was discharged with therapeutic subcutaneous enoxaparin. Two weeks after discharge, he was readmitted for anemia, at which time his skin lesion had improved significantly.
Discussion
Apixaban is reported to cause non-hemorrhagic adverse effects [6–9]. In leukocytoclastic vasculitis cases, lower-extremity skin purpura developed after 9 days of apixaban therapy [6,7]. It was managed with systemic steroid with complete resolution of skin changes within a few weeks [6,7]. In one case, the patient was successfully switched to rivaroxaban without any adverse events [7]. Veliyev et al. described a case of palmoplantar psoriasiform drug eruption in which cutaneous drug eruption developed after 3 days of apixaban therapy; it was managed with topical steroids, and skin changes resolved entirely within a few weeks [8]. Patil et al. reported a case of apixaban-induced delayed drug reaction-lichenoid reaction in which the patient was managed with topical steroids [9]. Similar to apixaban, rivaroxaban, another DOAC, has also been reported to cause skin toxicity. Soliman et al. reported a case of rivaroxaban-induced skin necrosis in which skin changes developed after 3 days of therapy; it was managed with a topical steroid, and the patient was successfully switched to low-molecular-weight heparin [10].
The pathophysiology behind DOAC-induced skin toxicity is unclear. Possible mechanisms include microvascular thrombosis, an imbalance between anticoagulant and procoagulant factors, autoimmune response, and direct medication toxicity [10]. The pathophysiology of warfarin-induced skin necrosis is well explained. Warfarin inactivates vitamin K-dependent coagulation factors II, VII, IX, and X. However, it also inactivates the vitamin K-dependent anticoagulants protein C and S. This may generate an imbalance between pro-coagulation and anticoagulation, which can lead to microvascular thrombi and skin necrosis [11,12]. It is unknown at this time if warfarin-induced skin necrosis and DOAC-induced skin toxicity share a similar pathophysiology.
There are some limitations to this case report. An autoimmune workup for drug-induced vasculitis was not performed. Protein C and Protein S levels could not be obtained because of inpatient restrictions as per hospital policy. A skin biopsy was not performed as the patient was on aspirin and Plavix as a home medication due to his significant coronary artery disease, and he was at high risk for bleeding and post-biopsy skin infections.
Our experience and literature review suggest that treatment of this condition with prompt discontinuation of the offending DOAC and conservative management with wound care is sufficient to manage the skin necrosis.
Conclusions
Although skin toxicities are recognized adverse effects of the new DOACs, including apixaban, they are rare and are usually associated with vasculitis. To the best of our knowledge, apixaban-induced skin necrosis has not been reported. This report has shown that skin toxicity can be associated with apixaban and that with the increasing use of these newer DOACs, clinicians should be aware of these potential adverse effects. In this case, the mainstay of treatment was the discontinuation of the causative agent and supportive treatment.
Disclaimer
This whole case report was supported by HCA Health Care-Kingwood Medical Center. The views expressed in this publication represent those of the authors and do not necessarily represent the official views of HCA Healthcare or any of its affiliated entities.
Conflict of interest
None
Abbreviations
DOACsdirect oral anticoagulants;
COPDchronic obstructive pulmonary disease;
CADcoronary artery disease;
CABGcoronary artery bypass graft;
CKDchronic kidney disease;
ProBNPpro brain natriuretic peptide;
V/Q scanventilation-perfusion scan;
AISNapixaban-induced skin necrosis
Figure 1. The left forearm shows a confluent and extensive area of skin necrosis with no raised skin lesions, no vesicles, and no purulent exudate. | Subcutaneous | DrugAdministrationRoute | CC BY-NC-ND | 33493142 | 18,870,085 | 2021-01-25 |
What was the outcome of reaction 'Skin necrosis'? | A 78-Year-Old Man with a Pulmonary Embolism Who Developed Skin Necrosis 7 Days After Treatment with the Direct Oral Anticoagulant Factor Xa Inhibitor Apixaban.
BACKGROUND Apixaban is one of the newer direct oral anticoagulants (DOACs) being used to manage venous thrombosis. Skin toxicities are recognized adverse effects of the new DOACs, but are rare and usually associated with vasculitis. This report is of a 78-year-old man admitted to the hospital with pulmonary thromboembolism, who developed severe and extensive skin necrosis of both forearms 7 days after treatment with apixaban. CASE REPORT A 78-year-old man was admitted for pulmonary embolism and congestive heart failure exacerbation. He was started on therapeutic enoxaparin and diuresis. Later on, enoxaparin was substituted with apixaban. Seven days after starting apixaban, he suddenly developed skin changes that developed into skin necrosis on both forearms and the abdominal wall. A skin biopsy was not performed due to the high risk of bleeding. Skin necrosis was diagnosed based on clinical findings. A review of clinical data and the patient's medication profile did not reveal any other possible etiology or culprit medication. Clinical presentation and lab values were not consistent with infections or autoimmune etiologies. Apixaban was discontinued as it was perceived to be the likely cause of skin necrosis. The skin changes gradually improved within 1 week with supportive wound care, and the patient did not require a skin graft. The patient was discharged safely with subcutaneous low-molecular-weight heparin therapy. CONCLUSIONS This report shows that skin toxicity can be associated with apixaban and that with the increasing use of these newer DOACs, clinicians should be aware of these potential adverse effects.
Background
In late 2010, a new class of anticoagulants known as direct oral anticoagulants (DOAC) was introduced as an alternative to conventional anticoagulation therapy [1]. DOACs have proven to be equal or superior to conventional therapy for preventing stroke in nonvalvular atrial fibrillation, with similar efficacy for prevention and treatment of venous thromboembolism [1,2]. This new class has clear advantages over warfarin, with better pharmacodynamics and pharmacokinetic properties, less drug-drug interactions, and no requirement for frequent coagulation test monitoring [2,3]. Bleeding risk is a significant concern for all anticoagulant therapies, but trials showed that compared to warfarin, DOACs have similar or less risk of major bleeding [4,5]. DOACs include direct thrombin inhibitors (Dabigatran) and direct factor Xa inhibitors (rivaroxaban, apixaban, and edoxaban).
There are reported data available for non-hemorrhagic adverse events caused by DOACs, such as leukocytoclastic vasculitis, psoriatic-like skin reaction, and lichenoid reaction [6–9]. Although skin toxicities are recognized adverse effects of the new DOACs, including apixaban, they are rare and are usually associated with vasculitis. This report is of a 78-year-old man admitted to the hospital with pulmonary thromboembolism, who developed severe and extensive skin necrosis of both forearms 7 days after treatment with apixaban.
Case Report
A 78-year-old man with a history of hypertension, chronic obstructive pulmonary disease (COPD), coronary artery disease (CAD), coronary artery bypass graft (CABG), diastolic heart failure, and stage 3 chronic kidney disease (CKD) presented with worsening dyspnea, swelling of the left leg, and pain for the last 2 days. On physical examination, he had bilateral basilar crackles with expiratory wheezing and bilateral lower-extremity edema. Troponin was 0.02 ng/ml, pro brain natriuretic peptide (ProBNP) was 1140 pg/ml, and D-dimer was elevated to 1066 ng/ml FEU. A chest X-ray showed right basilar lung infiltrate. Initially, he was started on bronchodilators for his COPD, diuresis for his congestive heart failure (CHF), and intravenous antibiotics for possible pneumonia. Pulmonary embolism was also considered in the differential diagnosis due to his increased oxygen requirement. CT angiography of the chest was not performed due to his elevated creatinine. A ventilation-perfusion lung scan (V/Q scan) showed an intermediate probability of pulmonary embolism. Ultrasound Doppler imaging of the lower extremities did not show deep vein thrombosis. The combination of the patient’s acute clinical presentation, elevated D-dimer, and V/Q scan intermediate probability was suggestive of pulmonary embolism. Therapeutic subcutaneous enoxaparin was initiated based on the creatinine clearance, as recommended by the pharmacy.
During the hospital stay, his condition was stabilized with bronchodilators and systemic glucocorticoids for advanced COPD, piperacillin/tazobactam for pneumonia, and diuresis for chronic diastolic congestive heart failure. His respiration was supported with a continued nasal cannula and BiPAP as needed. After 2 days of enoxaparin, he was switched to oral apixaban for his pulmonary embolism. His shortness of breath was improving slowly due to his underlying advanced COPD and CHF. During this time, he was waiting for a skilled nursing home placement.
After the seventh day of initiating apixaban, he developed a painful small ecchymosis on the left forearm, which quickly progressed to skin necrosis across the left forearm within 2 days (see Figure 1). He also developed skin necrosis on the right arm and the abdominal wall area. Surgery and Dermatology services were consulted. A skin biopsy was not performed due to the high risk of bleeding. The clinical presentation was not consistent with any autoimmune process. Apixaban was discontinued immediately. After reviewing all the patient’s medications during that interval, the clinical conclusion was a diagnosis of apixaban-induced skin necrosis. The skin changes gradually improved within 7 days with wound care and dressing. The patient did not require skin grafting, and he was discharged with therapeutic subcutaneous enoxaparin. Two weeks after discharge, he was readmitted for anemia, at which time his skin lesion had improved significantly.
Discussion
Apixaban is reported to cause non-hemorrhagic adverse effects [6–9]. In leukocytoclastic vasculitis cases, lower-extremity skin purpura developed after 9 days of apixaban therapy [6,7]. It was managed with systemic steroid with complete resolution of skin changes within a few weeks [6,7]. In one case, the patient was successfully switched to rivaroxaban without any adverse events [7]. Veliyev et al. described a case of palmoplantar psoriasiform drug eruption in which cutaneous drug eruption developed after 3 days of apixaban therapy; it was managed with topical steroids, and skin changes resolved entirely within a few weeks [8]. Patil et al. reported a case of apixaban-induced delayed drug reaction-lichenoid reaction in which the patient was managed with topical steroids [9]. Similar to apixaban, rivaroxaban, another DOAC, has also been reported to cause skin toxicity. Soliman et al. reported a case of rivaroxaban-induced skin necrosis in which skin changes developed after 3 days of therapy; it was managed with a topical steroid, and the patient was successfully switched to low-molecular-weight heparin [10].
The pathophysiology behind DOAC-induced skin toxicity is unclear. Possible mechanisms include microvascular thrombosis, an imbalance between anticoagulant and procoagulant factors, autoimmune response, and direct medication toxicity [10]. The pathophysiology of warfarin-induced skin necrosis is well explained. Warfarin inactivates vitamin K-dependent coagulation factors II, VII, IX, and X. However, it also inactivates the vitamin K-dependent anticoagulants protein C and S. This may generate an imbalance between pro-coagulation and anticoagulation, which can lead to microvascular thrombi and skin necrosis [11,12]. It is unknown at this time if warfarin-induced skin necrosis and DOAC-induced skin toxicity share a similar pathophysiology.
There are some limitations to this case report. An autoimmune workup for drug-induced vasculitis was not performed. Protein C and Protein S levels could not be obtained because of inpatient restrictions as per hospital policy. A skin biopsy was not performed as the patient was on aspirin and Plavix as a home medication due to his significant coronary artery disease, and he was at high risk for bleeding and post-biopsy skin infections.
Our experience and literature review suggest that treatment of this condition with prompt discontinuation of the offending DOAC and conservative management with wound care is sufficient to manage the skin necrosis.
Conclusions
Although skin toxicities are recognized adverse effects of the new DOACs, including apixaban, they are rare and are usually associated with vasculitis. To the best of our knowledge, apixaban-induced skin necrosis has not been reported. This report has shown that skin toxicity can be associated with apixaban and that with the increasing use of these newer DOACs, clinicians should be aware of these potential adverse effects. In this case, the mainstay of treatment was the discontinuation of the causative agent and supportive treatment.
Disclaimer
This whole case report was supported by HCA Health Care-Kingwood Medical Center. The views expressed in this publication represent those of the authors and do not necessarily represent the official views of HCA Healthcare or any of its affiliated entities.
Conflict of interest
None
Abbreviations
DOACsdirect oral anticoagulants;
COPDchronic obstructive pulmonary disease;
CADcoronary artery disease;
CABGcoronary artery bypass graft;
CKDchronic kidney disease;
ProBNPpro brain natriuretic peptide;
V/Q scanventilation-perfusion scan;
AISNapixaban-induced skin necrosis
Figure 1. The left forearm shows a confluent and extensive area of skin necrosis with no raised skin lesions, no vesicles, and no purulent exudate. | Recovering | ReactionOutcome | CC BY-NC-ND | 33493142 | 18,897,969 | 2021-01-25 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Intentional product use issue'. | Macular edema after rhegmatogenous retinal detachment repair: risk factors, OCT analysis, and treatment responses.
OBJECTIVE
To investigate risk factors, imaging characteristics, and treatment responses of cystoid macular edema (CME) after rhegmatogenous retinal detachment (RRD) repair.
METHODS
Consecutive, retrospective case-control series of patients who underwent pars plana vitrectomy (PPV) and/or scleral buckling (SB) for RRD, with at least six months of follow-up. Clinical and surgical parameters of patients with and without CME (nCME), based on spectral-domain optical coherence tomography (OCT), were compared.
RESULTS
Of 99 eyes enrolled, 25 had CME while 74 had nCME. Patients with CME underwent greater numbers of surgeries (P < 0.0001). After adjusting for number of surgeries, macula-off RRD (P = 0.06), proliferative vitreoretinopathy (PVR) (P = 0.09), surgical approach (PPV and/or SB, P = 0.21), and tamponade type (P = 0.10) were not statistically significant, although they all achieved significance on univariate analysis (P = 0.001 or less). Intraoperative retinectomy (P = 0.009) and postoperative pseudophakia or aphakia (P = 0.008) were more frequent in the CME group, even after adjustment. Characteristics of cCME on OCT included diffuse distribution, confluent cysts, and absence of subretinal fluid or intraretinal hyperreflective foci. Macular thickness improved significantly with intravitreal triamcinolone (P = 0.016), but not with anti-vascular endothelial growth factor agents (P = 0.828) or dexamethasone implant (P = 0.125). After adjusting for number of surgeries and macular detachment, final visual acuities remained significantly lower in the CME vs nCME group (P = 0.012).
CONCLUSIONS
Risk factors of CME include complex retinal detachment repairs requiring multiple surgeries, and pseudophakic or aphakic lens status. Although this cCME was associated with poor therapeutic response, corticosteroids were the most effective studied treatments.
Background
Cystoid macular edema (CME) is a common retinal condition characterized by macular thickening with intra-retinal fluid accumulation, often accompanied by decreased visual acuity (VA) [1]. It may develop as a complication of a wide spectrum of retinal diseases including diabetic retinopathy (DR), uveitis, exudative age-related macular degeneration (AMD), retinal vein occlusion (RVO), and genetic syndromes such as retinitis pigmentosa (RP) [2].
Although the pathophysiology of CME is multifactorial, breakdown of the inner blood retinal barrier is a common endpoint in most cases [1]. Current theories suggest subclinical inflammation as responsible for post-rhegmatogenous retinal detachment (RRD) CME [3]. While progressive leakage may be outlined with fluorescein angiography (FA) as the gold-standard for CME diagnosis, optical coherence tomography (OCT) is currently the most common imaging modality in the diagnosis and characterization of CME, as it is non-invasive and provides high resolution cross-sectional imaging of retinal anatomy [4], allowing easier and more frequent follow-up,
Rhegmatogenous retinal detachment is characterized by progressive accumulation of subretinal fluid due to retinal breaks. Although surgical repairs, including scleral buckle (SB) and pars plana vitrectomy (PPV), are effective surgical treatments, some cases with successful reattachment may have poor visual outcomes related to postoperative CME development [5, 6], which may persist for years in a minority of patients [7]. Retrospective and observational studies using FA and OCT have shown rates of post-vitrectomy CME varying from 5.5% after PPV for symptomatic floaters to 40% after complicated detachment repairs [6, 8, 9]. Treatments for CME primarily target inflammatory and pro-angiogenic mediators, but standard therapies such as anti-vascular endothelial growth factor (anti-VEGF) therapies may be ineffective for post-RRD CME [9, 10].
There is little data on post-RRD CME risk factors, rates, and anatomical characteristics [3, 5, 11]. Therefore, this observational study was designed to compare a consecutive case series of eyes with versus without post-RRD CME, with the aim to determine its risk factors and describe its clinical characteristics and therapeutic outcomes.
Methods
This was a retrospective, observational study approved by the medical center’s institutional review board, University of California Los Angeles Office of Human Research Protection (IRB#16-000574). This study adhered to the tenets of the Declaration of Helsinki and the rules of the Health Insurance Portability and Accountability Act of 1996.
Electronic health records (EHR) from a large academic referral center (Stein Eye Institute at UCLA) were reviewed. Current Procedural Terminology (CPT) coding records of surgical procedures from January 2015 to December 2017 were queried.
Population
All candidates underwent SB, PPV, or combined procedures for RRD, performed by two experienced vitreoretinal surgeons (JPH and SDS), with at least 6 months of follow up after surgery. Records were evaluated through July 2018.
Exclusion criteria were severe ocular trauma, uveitis, DR, endophthalmitis, RVO, myopic retinoschisis, or advanced dry or wet AMD.
Spectral Domain-OCT Analysis
All patients diagnosed with CME were examined with eye-tracked OCT. All OCTs were acquired with the Spectralis® (Heidelberg Engineering GmbH, Heidelberg, Germany) and RS-3000 (Nidek® Inc, San Jose, CA) devices. All CME was analyzed with Spectralis® OCTs consisting of 19 horizontal B-scans and manually adjusted for foveal centration. All OCT scans were carefully reviewed independently by two graders (CP, JPH) on the Heidelberg Eye Explorer software (Version 1.10.0.0).
A diagnosis of CME was noted if intraretinal hyporeflective spaces were noted in the inner nuclear layer (INL) and/or outer plexiform layer (OPL). Retinal thickness measurements were not used for CME diagnosis, as eyes had varying levels of atrophy.
Eyes were classified as having postoperative transient CME (tCME), chronic CME (cCME), or no CME (nCME). Both tCME and cCME were included as all CME (aCME) for statistical analysis. Postoperative tCME was defined as CME seen on OCT within 6 months of the final RRD, lasting less than 6months, and resolving using topical treatment. Postoperative cCME was defined as CME seen on two OCTs at least 6 months apart, based on previous reports [12].
Recorded characteristics of cCME on OCTs included presence of subretinal fluid, layers of CME involvement, presence of intraretinal hyperreflective foci, and integrity of outer retinal layers. Efficacy of anti-VEGF, triamcinolone acetonide (TA), or dexamethasone implant (Ozurdex®, Allergan Inc, Irvine, California) (DEX) injections were assessed after 4–6 weeks, if OCT was available. To determine treatment effect, pre- and post-injection OCTs were analyzed for central subfield thicknesses (CST) and inner macular volumes, comprised of the central five areas of the standard early treatment for diabetic retinopathy study (ETDRS) subfields [13].
Clinical charts analysis
Preoperative RRD parameters, intraoperative and post-operative data were collected. Glaucoma was counted if the patient carried this diagnose from a glaucoma specialist. Visual acuity was measured on a Snellen chart and converted to logarithm of the minimum angle of resolution (LogMAR) values for statistical analysis. Count fingers and hand motions vision were recorded as 1.98 and 2.28 LogMAR, respectively, based on previous studies using the Freiburg Visual Acuity Test [14]. Type of cCME treatment and number of intravitreal injections were included.
Statistical analysis
Qualitative values were listed as ratios and percentages while quantitative values were presented as mean ± standard deviation (SD). Qualitative variables were compared using the Fisher exact test. To compare continuous data between two groups, a Mann–Whitney U test was used. The Wilcoxon signed rank test was used to analyze changes in CST and inner retinal volume. The Kruskal–Wallis test was used to compare pre-injection OCT parameters between groups. The Shapiro–Wilk test assessed the normality of variable distribution. Covariate adjusted differences between CME groups were assessed using regression modeling (i.e. logistic, linear, and multinomial) using the number of surgeries as the covariate. Final visual acuity (logMAR) was log transformed in multivariable analyses and used the additional covariate of macula on/off. All statistics were performed in Stata SE 15.1 (StataCorp LP, College Station, TX). A P value of less than 0.05 was considered statistically significant. Denominators of ratios were less than the total number of eyes in the category if eyes could not be included in analyses due to missing or incomplete records.
Results
Population
A flowchart of population selection is shown in Fig. 1. A total of 508 surgical records were retrieved using CPT codes from January 2015 to December 2017. Of these, 133 eyes undergoing RRD repair met inclusion and exclusion criteria. Of these, 34 had less than 6 months of follow-up. The remaining 99 eyes of 97 patients were included for analysis. Of these, 20 patients (20%) had cCME, 5 (5%) had tCME, and 74 (75%) had nCME. Our primary analyses examine tCME and cCME as a single group, all CME (aCME), in comparison to nCME due to the small sample size for tCME. Descriptive statistics for all three groups can be found in the Additional file 1: Table S1.Fig. 1 Flowchart of patient selection process. ICD-9: International Classification of Disease, 9th edition. CPT Current Procedural Terminology, CME Cystoid Macular Edema
CME risk factors
Demographic and surgical data are summarized by CME group in Table 1. There was no difference in age at last surgery between patients in the aCME group (64.1 ± 11.6 years) versus patients in the nCME group (56.7 ± 18.0 years, P = 0.092). There was no significant difference in gender (P = 0.093), glaucoma status (P = 0.258), or length of follow-up (P = 0.869). Among those with glaucoma, there was no difference in the rates of topical prostaglandin analogs, other topical medications, or glaucoma surgery between groups (P = 0.992).Table 1 Demographics, baseline characteristics, and surgical data of patients with aCME and nCME
aCME nCME P value Adjusted P value1
Demographic data
Number of eyes 25 (25%) 74 (75%)
Follow-up (months) 21.4 ± 12.1 20.4 ± 10.8 0.87
Sex, Female 8 (32%) 38 (34%) 0.09
Age (years) 64.1 ± 11.6 56.7 ± 18.0 0.09
Clinical data
Right eye 11 (44%) 40 (54%) 0.38
Glaucoma 4 (16%) 6 (8%) 0.26
Lens status < 0.001 0.008
Phakic 1 (4%) 44 (60%)
Pseudophakic 14 (56%) 28 (38%)
Aphakic 10 (40%) 2 (3%)
Macula offa 20/24 (83%) 31/70 (44%) 0.001 0.06
PVR Stage Ca 15/24 (63%) 5/74 (7%) < 0.001 0.09
Final VA (LogMAR) 0.85 ± 0.80 0.20 ± 0.30 < 0.001 0.012b
ERM 18 (72%) 28 (38%) 0.005
Surgical details
Number of surgeries 3.5 ± 1.8 1.4 ± 0.9 < 0.001
Multiple PPV 21 (84%) 17 (23%) < 0.001 –c
Referred after surgery elsewhere 12 (48%) 5 (7%) < 0.001 0.31
Number of surgery outside 1 ± 1.3 0.095 ± 0.4 < 0.001
Type of surgery < 0.001 0.21
SB 1 (4%) 25 (34%)
PPV 7 (28%) 28 (38%)
PPV + SB 17 (68%) 21 (28%)
Tamponade agent < 0.001 0.10
None/Air 1 (4%) 24 (32%)
Gas (SF6 or C3F8) 8 (32%) 46 (62%)
Silicone oil 16 (64%) 4 (5%)
Cryotherapya 4/24 (17%) 30/73 (41%) 0.047 0.036
Retinectomy 9 (36%) 4 (5%) < 0.001 0.009
PFCLa 18/23 (78%) 35/47 (75%) 0.73 0.38
aCME all (chronic + transient) cystoid macular edema, nCME no cystoid macular edema, PVR proliferative vitreoretinopathy, VA visual acuity, LogMAR (logarithm of the minimum angle of resolution), PPV pars-plana vitrectomy, ERM epiretinal membrane, SB scleral buckle, PFCL perfluorocarbon liquid
aDenominators are provided if the number is less than the total number of eyes in the category due to missing or incomplete data
bFinal VA adjusted P value from a model with covariates for total number of surgeries and Macula on/off
cAdjusted model not possible due to collinearity of Multiple PPV with number of surgeries (i.e. those with Multiple PPV had greater than 2 surgeries, while those with no PPV had fewer)
1P-value for difference after adjustment for total number of surgeries
Eyes in the aCME group underwent a significantly greater number of retinal surgeries (3.5 ± 1.8) compared with eyes in the nCME group (1.4 ± 1.9) (P < 0.001). Due to the high collinearity between CME status and number of surgeries, multivariate analysis using this as a covariate was performed. Final lens status differed significantly between groups after adjustment (P = 0.008), with only one eye in the aCME group remaining phakic. A higher rate of aCME eyes had a macula-off retinal detachment (20/24, 83%), compared with nCME eyes (31/70, 44%, P = 0.001). Proliferative vitreoretinopathy (PVR) stage C was more frequent in the aCME group (15/24, 63%) versus the nCME group (5/74, 7%), P < 0.0001. However, both macula-off status (P = 0.06) and presence of PVR C (P = 0.09) lost statistical significance after adjustment for the total number of surgeries performed. Surgical approaches were statistically different between the aCME and nCME groups: primary SB in 1/25 (4%) aCME eyes vs. 25/74 (34%) nCME eyes, PPV in 7/25 (28%) aCME eyes vs. 28/74 (38%) nCME eyes, and combined SB + PPV in 17/25 (68%) aCME eyes vs. 21/74 (28%) nCME eyes (P <0.0001). However, these differences in the surgical approach were not reliably different after adjustment for the number of surgeries. Rates of retinectomy were higher in the aCME group than the nCME group after adjustment (9/25, 36% vs 4/74, 5%, P = 0.009). Rates of cryotherapy were higher in the nCME group (30/74, 41%) than aCME group (4/24, 17%), even after adjustment (P = 0.036). Unadjusted differences in tamponade agent between groups were statistically significant (P < 0.0001). Notably, 16 out of 25 (64%) aCME eyes received silicone oil (SO) at least once, while only 4 out of 74 (5%) of nCME eyes did. However, tamponade differences were no long significant after covariate adjustment. There was no difference in the use of perfluorocarbon liquid (PFCL) (P = 0.728).
At last examination, VA was significantly lower in aCME group (0.85 ± 0.80 LogMAR) than in nCME group (0.20 ± 0.30 LogMAR), P < 0.0001. When adjusting for the number of surgeries and macular detachment, the marginal estimates for between group differences in LogMAR were attenuated (aCME = 0.55 vs nCME = 0.26), though still statistically significant (P = 0.012).
Two patients had non-simultaneous RRDs in each eye. One patient was 23 years of age at the time of both surgeries and underwent SB with cryotherapy in each eye for inferior chronic RRD, without CME development. The other patient was 83 at the time of final surgery in both eyes, had initial surgeries performed elsewhere, had multiple PPVs in both eyes, and received SO in both eyes, and this patient developed cCME in both eye.
OCT characteristics of cCME
Eyes in the cCME group (n = 20) shared particular qualities on OCT (Fig. 2). All eyes had diffuse CME involving the four macular quadrants. The CME always involved the fovea but had variable extent into peripheral macula and was often asymmetric. Cysts were uniformly present in the INL and OPL, with occasional ganglion cell layer involvement. Florid CME often assumed a retinoschitic appearance. With time, cysts coalesced into larger confluent cavities with irregular, polygonal shapes. These cysts often spanned within the same retinal layer and across adjacent layers. Temporary resolution of these cysts after treatment disclosed disorganization and variable atrophy of the retinal layers in areas of cyst confluency. If CME recurred after treatment, it typically recurred in the same distribution of the macula.Fig. 2 Spectral-domain optical coherence tomography and infrared image elevation overlays of two different patients with chronic cystoid macular edema post-rhegmatogenous retinal detachment. The scan in Row A demonstrates schisis-like changes. The scan in Row B demonstrates confluent cystic cavities spanning retinal layers that developed over two years. In both scans, note diffuse, asymmetric distribution of retinal cysts crossing the horizontal raphe, involvement of inner and outer retinal layers, absence of subretinal fluid, and relative preservation of outer retinal bands subjacent to retinal edema
Outer retinal layer integrity was heterogeneous. On the first OCT with CME after the final RRD repair, ellipsoid zone (EZ) disruption was seen in 18 eyes (90%), external limiting membrane (ELM) disruption in 14 eyes (80%), and retinal pigment epithelial (RPE) disruption in 11 eyes (55%). Remarkably, there was no case with subretinal fluid (SRF), and no case of intraretinal hyperreflective foci or hemorrhage.
An epiretinal membrane (ERM) was detectable on OCT during the post-operative follow-up period in 17/20 (85%) cCME eyes, 2/5 (40%) tCME eyes, and 28/74 (38%) of nCME eyes (P = 0.005). Evidence of traction on OCT, such as inner retinal wrinkling or ectopic inner foveal layers, was appreciable in only 4 of the 17 cCME eyes with ERM. However, the severity of CME was out of proportion to the ERM changes in all but one of these four eyes.
CME Treatments
All patients with tCME (n = 5) and cCME (n = 20) received topical medications. Intravitreal injections and surgical interventions were administered according to physician discretion. All patients received corticosteroid drops, non-steroidal anti-inflammatory agent (NSAID) drop, or a combination of both for at least two months after the diagnosis of CME. If the CME failed to respond, patients thereafter received intravitreal injections of anti-VEGF (bevacizumab, ranibizumab, aflibercept), or steroids (triamcinolone acetate (TA), and/or dexamethasone intravitreal implant (DEX)).
The five patients (25%) with tCME had permanent resolution of CME with drops. Table 2 summarizes intravitreal treatments and anatomical responses of cCME. Five patients received at least one bevacizumab (Avastin®, Genentech Inc., San Francisco, CA, USA) injection, and one of these patients also received aflibercept (Eylea®, Regeneron Inc., Tarrytown, NY, USA) injections. In cCME eyes, there was a significant CST (P = 0.016, Wilcoxon signed rank test) and volume (P = 0.016) decrease after TA. (P = 0.125) (Fig. 3). There was no difference in pre-injection CST or volume between groups (P = 0.397, P = 0.457). There was no significant change in CST or volume with anti-VEGF treatment (P =0.915, P = 0.828) or DEX (P = 0.434, P = 0.125). No patient developed elevated intraocular pressure (IOP) after intravitreal injection requiring treatment. One patient developed sterile endophthalmitis after her seventh TA injection that spontaneously resolved without sequelae. A PPV for an ERM was performed in 9/16 cCME eyes with OCT evidence of ERM, with full resolution of the CME in only one eye.Table 2 Treatments for chronic cystoid macular edema (cCME) and anatomical responses on spectral-domain optical coherence tomography
Type of treatment Anti-VEGF TA DEX
Number of eyes 5 7 4
Number of Injections (Median; [Range]) 2.5, 1-14 2.0, 1-10 2.5, 1-7
CST pre-injection (μm) 401 ± 84.9 481 ± 104 397 ± 57.0 P = 0.397
CST post-injection (μm) 393 ± 106 402 ± 102 355 ± 80.4
Percent CST change (μm) − 1.44 ± 17.1, P = 0.915 − 15.6 ± 16.6, P = 0.016 − 11.0 ± 10.7, P = 0.434
Inner macular volume pre-injection (mm3) 2.81 ± 0.43 3.18 ± 0.56 3.12 ± 0.80 P = 0.457
Inner macular volume post-injection (mm3) 2.74 ± 0.53 2.72 ± 0.53 2.66 ± 0.486
Percent (%) inner macular volume change (mm3) −2.49 ± 12.35, P = 0.828 −13.9 ± 10.8, P = 0.016 −10.7 ± 25.7, P = 0.125
Values are listed as averages with standard deviations
VEGF vascular endothelial growth factor, TA triamcinolone acetate, DEX dexamethasone implant, CST central subfield thickness
Fig. 3 Spectral-domain optical coherence tomography (OCT) images of chronic cystoid macular edema (CME) post-rhegmatogenous retinal detachment (RRD) repair of the left eye, with dates and visual acuities (VA). Panel A: OCT prior to dexamethasone implant (DEX) injection. Panel B: OCT 1 month after DEX injection, showing resolution of CME but retinal layer atrophy. Modest VA improvement was noted. Panel C: OCT four months after injection, showing recurrence of CME in a similar distribution and slight decrease in VA
Discussion
Chronic CME after retinal detachment repair remains a challenging complication. In this paper, the risk factors for post-RRD CME, its OCT characteristics, and treatments outcomes are described.
Chronic post-RRD CME is thought to be pathophysiologically distinct from other etiologies of CME [3]. Among CME etiologies such as uveitis, RVO, and DME, many of the cytokines and damaged tissue responses are shared [1, 2, 15]. Certain CME etiologies, however, may have unique pathophysiologic mechanisms despite phenotypic similarities [16]. Entities with a significant pro-angiogenic component, such as exudative AMD, may respond to anti-VEGF agents, while those with a broad inflammatory component, such as uveitic CME or Irvine-Gass syndrome, may respond better to anti-inflammatory drugs [12].
While some studies found no risk factor differences for CME rates [5, 17], some series have, on univariate analyses, reported increased rates in pseudophakic [18] and aphakic eyes [6], older patients, more extensive RRD, and a history of a detached macula. In the present study, lens status was significantly different between groups, with increased pseudophakia and aphakia in aCME eyes. Unicameral communication in vitrectomized eyes modifies circulation of inflammatory cytokines, as animal studies have noted changes in oxygen and antioxidant gradients [19]. Higher rates of pseudophakia/aphakia in the aCME group may be related either to the actual lens surgery or to the complexity of the vitreo-retinal surgeries requiring lens extraction. As a substantial proportion of eyes with complicated RRD will be made pseudophakic or aphakic, anticipating CME in complex cases can have prognostic implications.
Eyes with CME had a greater number of surgeries, higher rates of PVR grade C and retinectomy, and higher rates of SO use. Many studies have shown increased inflammation and CME with more complicated ocular surgeries and inflammatory risk factors [3, 11, 20]. Re-detachments are frequently associated with PVR formation and warrant additional surgeries, both of which can increase intraocular inflammation and possible risk for CME [21]. Retinectomy is helpful when PVR membranes are not amenable to mechanical peeling, and therefore retinectomy likely indicates severe pathology rather than directly causing CME.
Macular detachment was associated with a higher risk of CME, which is in line with prior papers [18]. Of note, previous studies have noted outer nuclear layer CME on OCT of the detached macula [22, 23]. Although the retinal hydration theory, implicated in macular hole edema formation [24], may contribute to post-RRD CME, the presence of leakage on FA suggests dynamic fluid movements as opposed to static, non-leaking cysts. Moreover, absence of SRF after RD repair would theoretically lead to rapid elimination of intraretinal fluid by normal pumping mechanisms. Although such studies for macular detachment and CME development have not been explored [18, 25], permanent damage to retinal cellular elements while detached may lead to persistent dysfunction and contribute to CME.
There was a significant difference in surgical approaches between groups, with higher rates of combined SB and PPV in aCME eyes. This is not surprising, given that scleral buckles are often combined with PPV for complex or recurrent detachments to support the vitreous base and/or areas of retinal pathology. However, there was significantly more cryotherapy in the nCME group. Cryotherapy at our institution is only used during primary scleral buckling, usually for limited and uncomplicated detachments in phakic patients. While data comparing CME rates between PPV and SB are scant, the correlation between more complicated detachments and CME is consistent [3, 11, 18].
After adjusting for the number of surgeries, type of surgery (P = 0.21), macular detachment (P = 0.06), PVR Grade C (P = 0.09) and tamponade type (P = 0.10) lost statistical significance. This may be related to the limited sample size, as there remained a trend towards significance. Moreover, these factors are clinically related to the number of surgeries and surgical failure. The interplay of inflammation among these factors requires more formal study.
Characteristics of CME on OCT can be useful diagnostic clues, and post-RRD cCME displays distinguishing OCT features (Fig. 2). Previous studies have examined OCTs of various conditions associated with CME and noted distinctive findings [13]. These findings could then be used to diagnose conditions accurately as well as account for variability in VA [26]. Post-RRD cCME shares features of uveitic CME, such as diffuse macular distribution, inner and outer layer cysts, and absence of hyperreflective foci. This contrasts to post-RRD tCME, which is much less severe, more central and fleeting, and may be a variant of pseudophakic CME.
The presence of ERM is common after RRD and may confound CME diagnosis [16]. Although there was a significant difference between groups in the presence of ERM on OCT, there was resolution of CME in only one eye after ERM peeling, suggesting that traction plays a small role in most cases of post-RRD CME. Therefore, there should be high suspicion for post-RRD cCME in any patient status-post RRD repair that has severe, diffuse CME without SRF in the absence of other typical inflammatory or tractional signs.
The RPE has a well-studied role in pumping syneretic vitreous fluid through the retina and into the choroidal space [1]. Active fluid transport regulation by the RPE and Muller cells along with maintenance of tight junctional proteins are thought to mitigate CME accumulation [1, 2, 15], and dysfunction of these cells causes an imbalance of fluid inflow and egress. Previous papers examining CME OCT findings note varying SRF rates, from 5% in uveitic CME up to 100% in central RVO-associated CME [1, 4, 13, 27]. Therefore, the absence of SRF in cCME suggests a grossly functioning RPE and outer retinal barrier.
Intravitreal corticosteroids were more effective than intravitreal anti-VEGF or topical medications for cCME in our series. Recent investigations have shown success with intravitreal corticosteroids for chronic post-RRD CME [16, 25]. Thanos et al. found favorable responses to DEX all eyes, but in all cases CME recurred after 3 months. This aligns with pharmacokinetic studies showing a dual-phase response of high dexamethasone concentrations for the first 2 months after delivery followed by a precipitous decrease during the third month [28]. Experimental studies have demonstrated a reduced half-life of anti-VEGF agents and triamcinolone acetate in vitrectomized eyes compared with non-vitrectomized eyes [28, 29], but similar clearances between eyes with DEX. Statistically significance for anatomical improvement was not reached for DEX in our series, likely due to the small number of eyes. Moreover, aphakia has been suggested to cause increased unicameral circulation of inflammatory cytokines [6, 30], but aphakia precludes the use of DEX. One randomized controlled trial evaluating PPV with SO for RRD with grade C PVR found a significant decrease in CME occurrence at 6 months post-operatively in those with intraoperative DEX [31]. Corticosteroids have been shown to modulate a number of cytokines secreted by retinal cells, such as tumor necrosis factor-α, interleukins-1β, 6, and 8, as well as induce expression of occludin, ZO-1, and claudin-5 [1, 16, 31]. Steroids also modulate expression of aquaporin, predominantly expressed in end-feet of Müller cells and astrocytes. Corticosteroids may therefore stabilize the BRB and encourage resolution of CME, accounting for the increased efficacy of corticosteroids over anti-VEGF agents. Nevertheless, disadvantages of TA and DEX include accelerated cataract formation and risk of increased IOP; however, most patients with cCME will require cataract extraction, and no patient in our series required treatment for ocular hypertension.
Average final visit VA was significantly worse in the aCME group even after adjusting for macula-off status and number of surgeries. Reports on recalcitrant CME after PPV for RRD, despite anatomic improvement, found only short-term visual acuity gains [16, 25].
Irvine-Gass syndrome (IGS) is another potential diagnosis in these cases. We did not regularly perform FA or optic disc evaluations to check for optic nerve head leakage during the course of follow up. However, IGS is not described after PPV and has been described as a potential treatment option in many cases [32]. Therefore, IGS would have likely responded to topical treatments, steroid injections, or PPV. The OCT appearance of IGS is also less diffuse, more foveocentric, and may be associated with SRF, as opposed to characteristics noted with post-RRD cCME.
Our paper has a relatively large sample size of post-RRD CME, long-term patient records and follow up, and variety of treatments. Despite this, our study has several limitations. The retrospective analysis precluded standardized imaging and treatment protocols. Significant loss to follow-up likely led to underreporting of chronic post-RRD CME and an inability to accurately determine incidence. The high percentage of CME likely relates to inclusion of eyes that had initial RRD repairs prior to the inclusion period and multiple referrals for complex cases. We were unable to determine after which surgery CME appeared due to inconsistent timing and absence of OCT acquisition between surgeries, or missing outside records. A small number of eyes received anti-VEGF injections, and greater numbers may show CME improvement. A larger, prospective study evaluating complex macular surgeries is warranted.
In conclusion, cCME after RRD is a complex entity with interconnected risk factors. A high index of suspicion based on risk factor and imaging characteristics can allow anticipation of cCME development and early treatment. Currently, corticosteroids have the most evidence of treatment success, and prompt intervention may provide better functional and structural outcomes.
Supplementary information
Additional file 1: Table S1. Descriptive statistics for all three groups, as explained in results population section.
Abbreviations
CMECystoid macular edema
tCMETransient CME
cCMEChronic CME
aCMEAll CME
DRDiabetic retinopathy
AMDAge-related macular degeneration
RVORetinal vein occlusion
RPRetinitis pigmentosa
RRDRhegmatogenous retinal detachment
FAFluorescein angiography
OCTOptical coherence tomography
SBScleral buckle
PPVPars plana vitrectomy
VEGFVascular endothelial growth factor
EHRElectronic health records
CPTCurrent procedural terminology
TATriamcinolone acetonide
DEXDexamethasone intravitreal implant
CSTCentral subfield thickness
ETDRSEarly treatment for diabetic retinopathy study
SDStandard deviation
PVRProliferative vitreoretinopathy
PFCLPerfluorocarbon liquid
INLInner nuclear layer
OPLOuter plexiform layer
EZEllipsoid zone
ELMExternal limiting membrane
ERMEpiretinal membrane
RPERetinal pigment epithelium
SRFSubretinal fluid
NSAIDNon-steroidal anti-inflammatory drug
IOPIntraocular pressure
SOSilicone oil
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Supplementary information accompanies this paper at 10.1186/s40942-020-00254-9.
Acknowledgements
The authors would like to thank Nicholas J. Jackson, PhD, for his statistical assistance and valuable review of the manuscript.
Authors’ contributions
All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by CP, IC, AG, SG, SDS, and JPH. The first draft of the manuscript was written by CP, and all authors commented on draft versions of the manuscript. All authors read and approved the final manuscript.
Funding
The research described was supported by NIH/National Center for Advancing Translational Science (NCATS) UCLA CTSI Grant Number UL1TR001881, and by an unrestricted grant from Research to Prevent Blindness.
Availability of data and materials
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.
Ethics approval and consent to participate
This research study was conducted retrospectively from data obtained for clinical purposes. An IRB official waiver of ethical approval was granted from the IRB of the University of California Los Angeles Office of Human Research Protection (IRB#16-000574).
Consent for publication
Not applicable.
Competing interests
All authors certify that they have no affiliations with or involvement in any organization or entity with any financial interest (such as honoraria; educational grants; participation in speakers’ bureaus; membership, employment, consultancies, stock ownership, or other equity interest; and expert testimony or patent-licensing arrangements), or non-financial interest (such as personal or professional relationships, affiliations, knowledge or beliefs) in the subject matter or materials discussed in this manuscript. | TRIAMCINOLONE ACETONIDE | DrugsGivenReaction | CC BY | 33494835 | 18,924,022 | 2021-01-25 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Non-infectious endophthalmitis'. | Macular edema after rhegmatogenous retinal detachment repair: risk factors, OCT analysis, and treatment responses.
OBJECTIVE
To investigate risk factors, imaging characteristics, and treatment responses of cystoid macular edema (CME) after rhegmatogenous retinal detachment (RRD) repair.
METHODS
Consecutive, retrospective case-control series of patients who underwent pars plana vitrectomy (PPV) and/or scleral buckling (SB) for RRD, with at least six months of follow-up. Clinical and surgical parameters of patients with and without CME (nCME), based on spectral-domain optical coherence tomography (OCT), were compared.
RESULTS
Of 99 eyes enrolled, 25 had CME while 74 had nCME. Patients with CME underwent greater numbers of surgeries (P < 0.0001). After adjusting for number of surgeries, macula-off RRD (P = 0.06), proliferative vitreoretinopathy (PVR) (P = 0.09), surgical approach (PPV and/or SB, P = 0.21), and tamponade type (P = 0.10) were not statistically significant, although they all achieved significance on univariate analysis (P = 0.001 or less). Intraoperative retinectomy (P = 0.009) and postoperative pseudophakia or aphakia (P = 0.008) were more frequent in the CME group, even after adjustment. Characteristics of cCME on OCT included diffuse distribution, confluent cysts, and absence of subretinal fluid or intraretinal hyperreflective foci. Macular thickness improved significantly with intravitreal triamcinolone (P = 0.016), but not with anti-vascular endothelial growth factor agents (P = 0.828) or dexamethasone implant (P = 0.125). After adjusting for number of surgeries and macular detachment, final visual acuities remained significantly lower in the CME vs nCME group (P = 0.012).
CONCLUSIONS
Risk factors of CME include complex retinal detachment repairs requiring multiple surgeries, and pseudophakic or aphakic lens status. Although this cCME was associated with poor therapeutic response, corticosteroids were the most effective studied treatments.
Background
Cystoid macular edema (CME) is a common retinal condition characterized by macular thickening with intra-retinal fluid accumulation, often accompanied by decreased visual acuity (VA) [1]. It may develop as a complication of a wide spectrum of retinal diseases including diabetic retinopathy (DR), uveitis, exudative age-related macular degeneration (AMD), retinal vein occlusion (RVO), and genetic syndromes such as retinitis pigmentosa (RP) [2].
Although the pathophysiology of CME is multifactorial, breakdown of the inner blood retinal barrier is a common endpoint in most cases [1]. Current theories suggest subclinical inflammation as responsible for post-rhegmatogenous retinal detachment (RRD) CME [3]. While progressive leakage may be outlined with fluorescein angiography (FA) as the gold-standard for CME diagnosis, optical coherence tomography (OCT) is currently the most common imaging modality in the diagnosis and characterization of CME, as it is non-invasive and provides high resolution cross-sectional imaging of retinal anatomy [4], allowing easier and more frequent follow-up,
Rhegmatogenous retinal detachment is characterized by progressive accumulation of subretinal fluid due to retinal breaks. Although surgical repairs, including scleral buckle (SB) and pars plana vitrectomy (PPV), are effective surgical treatments, some cases with successful reattachment may have poor visual outcomes related to postoperative CME development [5, 6], which may persist for years in a minority of patients [7]. Retrospective and observational studies using FA and OCT have shown rates of post-vitrectomy CME varying from 5.5% after PPV for symptomatic floaters to 40% after complicated detachment repairs [6, 8, 9]. Treatments for CME primarily target inflammatory and pro-angiogenic mediators, but standard therapies such as anti-vascular endothelial growth factor (anti-VEGF) therapies may be ineffective for post-RRD CME [9, 10].
There is little data on post-RRD CME risk factors, rates, and anatomical characteristics [3, 5, 11]. Therefore, this observational study was designed to compare a consecutive case series of eyes with versus without post-RRD CME, with the aim to determine its risk factors and describe its clinical characteristics and therapeutic outcomes.
Methods
This was a retrospective, observational study approved by the medical center’s institutional review board, University of California Los Angeles Office of Human Research Protection (IRB#16-000574). This study adhered to the tenets of the Declaration of Helsinki and the rules of the Health Insurance Portability and Accountability Act of 1996.
Electronic health records (EHR) from a large academic referral center (Stein Eye Institute at UCLA) were reviewed. Current Procedural Terminology (CPT) coding records of surgical procedures from January 2015 to December 2017 were queried.
Population
All candidates underwent SB, PPV, or combined procedures for RRD, performed by two experienced vitreoretinal surgeons (JPH and SDS), with at least 6 months of follow up after surgery. Records were evaluated through July 2018.
Exclusion criteria were severe ocular trauma, uveitis, DR, endophthalmitis, RVO, myopic retinoschisis, or advanced dry or wet AMD.
Spectral Domain-OCT Analysis
All patients diagnosed with CME were examined with eye-tracked OCT. All OCTs were acquired with the Spectralis® (Heidelberg Engineering GmbH, Heidelberg, Germany) and RS-3000 (Nidek® Inc, San Jose, CA) devices. All CME was analyzed with Spectralis® OCTs consisting of 19 horizontal B-scans and manually adjusted for foveal centration. All OCT scans were carefully reviewed independently by two graders (CP, JPH) on the Heidelberg Eye Explorer software (Version 1.10.0.0).
A diagnosis of CME was noted if intraretinal hyporeflective spaces were noted in the inner nuclear layer (INL) and/or outer plexiform layer (OPL). Retinal thickness measurements were not used for CME diagnosis, as eyes had varying levels of atrophy.
Eyes were classified as having postoperative transient CME (tCME), chronic CME (cCME), or no CME (nCME). Both tCME and cCME were included as all CME (aCME) for statistical analysis. Postoperative tCME was defined as CME seen on OCT within 6 months of the final RRD, lasting less than 6months, and resolving using topical treatment. Postoperative cCME was defined as CME seen on two OCTs at least 6 months apart, based on previous reports [12].
Recorded characteristics of cCME on OCTs included presence of subretinal fluid, layers of CME involvement, presence of intraretinal hyperreflective foci, and integrity of outer retinal layers. Efficacy of anti-VEGF, triamcinolone acetonide (TA), or dexamethasone implant (Ozurdex®, Allergan Inc, Irvine, California) (DEX) injections were assessed after 4–6 weeks, if OCT was available. To determine treatment effect, pre- and post-injection OCTs were analyzed for central subfield thicknesses (CST) and inner macular volumes, comprised of the central five areas of the standard early treatment for diabetic retinopathy study (ETDRS) subfields [13].
Clinical charts analysis
Preoperative RRD parameters, intraoperative and post-operative data were collected. Glaucoma was counted if the patient carried this diagnose from a glaucoma specialist. Visual acuity was measured on a Snellen chart and converted to logarithm of the minimum angle of resolution (LogMAR) values for statistical analysis. Count fingers and hand motions vision were recorded as 1.98 and 2.28 LogMAR, respectively, based on previous studies using the Freiburg Visual Acuity Test [14]. Type of cCME treatment and number of intravitreal injections were included.
Statistical analysis
Qualitative values were listed as ratios and percentages while quantitative values were presented as mean ± standard deviation (SD). Qualitative variables were compared using the Fisher exact test. To compare continuous data between two groups, a Mann–Whitney U test was used. The Wilcoxon signed rank test was used to analyze changes in CST and inner retinal volume. The Kruskal–Wallis test was used to compare pre-injection OCT parameters between groups. The Shapiro–Wilk test assessed the normality of variable distribution. Covariate adjusted differences between CME groups were assessed using regression modeling (i.e. logistic, linear, and multinomial) using the number of surgeries as the covariate. Final visual acuity (logMAR) was log transformed in multivariable analyses and used the additional covariate of macula on/off. All statistics were performed in Stata SE 15.1 (StataCorp LP, College Station, TX). A P value of less than 0.05 was considered statistically significant. Denominators of ratios were less than the total number of eyes in the category if eyes could not be included in analyses due to missing or incomplete records.
Results
Population
A flowchart of population selection is shown in Fig. 1. A total of 508 surgical records were retrieved using CPT codes from January 2015 to December 2017. Of these, 133 eyes undergoing RRD repair met inclusion and exclusion criteria. Of these, 34 had less than 6 months of follow-up. The remaining 99 eyes of 97 patients were included for analysis. Of these, 20 patients (20%) had cCME, 5 (5%) had tCME, and 74 (75%) had nCME. Our primary analyses examine tCME and cCME as a single group, all CME (aCME), in comparison to nCME due to the small sample size for tCME. Descriptive statistics for all three groups can be found in the Additional file 1: Table S1.Fig. 1 Flowchart of patient selection process. ICD-9: International Classification of Disease, 9th edition. CPT Current Procedural Terminology, CME Cystoid Macular Edema
CME risk factors
Demographic and surgical data are summarized by CME group in Table 1. There was no difference in age at last surgery between patients in the aCME group (64.1 ± 11.6 years) versus patients in the nCME group (56.7 ± 18.0 years, P = 0.092). There was no significant difference in gender (P = 0.093), glaucoma status (P = 0.258), or length of follow-up (P = 0.869). Among those with glaucoma, there was no difference in the rates of topical prostaglandin analogs, other topical medications, or glaucoma surgery between groups (P = 0.992).Table 1 Demographics, baseline characteristics, and surgical data of patients with aCME and nCME
aCME nCME P value Adjusted P value1
Demographic data
Number of eyes 25 (25%) 74 (75%)
Follow-up (months) 21.4 ± 12.1 20.4 ± 10.8 0.87
Sex, Female 8 (32%) 38 (34%) 0.09
Age (years) 64.1 ± 11.6 56.7 ± 18.0 0.09
Clinical data
Right eye 11 (44%) 40 (54%) 0.38
Glaucoma 4 (16%) 6 (8%) 0.26
Lens status < 0.001 0.008
Phakic 1 (4%) 44 (60%)
Pseudophakic 14 (56%) 28 (38%)
Aphakic 10 (40%) 2 (3%)
Macula offa 20/24 (83%) 31/70 (44%) 0.001 0.06
PVR Stage Ca 15/24 (63%) 5/74 (7%) < 0.001 0.09
Final VA (LogMAR) 0.85 ± 0.80 0.20 ± 0.30 < 0.001 0.012b
ERM 18 (72%) 28 (38%) 0.005
Surgical details
Number of surgeries 3.5 ± 1.8 1.4 ± 0.9 < 0.001
Multiple PPV 21 (84%) 17 (23%) < 0.001 –c
Referred after surgery elsewhere 12 (48%) 5 (7%) < 0.001 0.31
Number of surgery outside 1 ± 1.3 0.095 ± 0.4 < 0.001
Type of surgery < 0.001 0.21
SB 1 (4%) 25 (34%)
PPV 7 (28%) 28 (38%)
PPV + SB 17 (68%) 21 (28%)
Tamponade agent < 0.001 0.10
None/Air 1 (4%) 24 (32%)
Gas (SF6 or C3F8) 8 (32%) 46 (62%)
Silicone oil 16 (64%) 4 (5%)
Cryotherapya 4/24 (17%) 30/73 (41%) 0.047 0.036
Retinectomy 9 (36%) 4 (5%) < 0.001 0.009
PFCLa 18/23 (78%) 35/47 (75%) 0.73 0.38
aCME all (chronic + transient) cystoid macular edema, nCME no cystoid macular edema, PVR proliferative vitreoretinopathy, VA visual acuity, LogMAR (logarithm of the minimum angle of resolution), PPV pars-plana vitrectomy, ERM epiretinal membrane, SB scleral buckle, PFCL perfluorocarbon liquid
aDenominators are provided if the number is less than the total number of eyes in the category due to missing or incomplete data
bFinal VA adjusted P value from a model with covariates for total number of surgeries and Macula on/off
cAdjusted model not possible due to collinearity of Multiple PPV with number of surgeries (i.e. those with Multiple PPV had greater than 2 surgeries, while those with no PPV had fewer)
1P-value for difference after adjustment for total number of surgeries
Eyes in the aCME group underwent a significantly greater number of retinal surgeries (3.5 ± 1.8) compared with eyes in the nCME group (1.4 ± 1.9) (P < 0.001). Due to the high collinearity between CME status and number of surgeries, multivariate analysis using this as a covariate was performed. Final lens status differed significantly between groups after adjustment (P = 0.008), with only one eye in the aCME group remaining phakic. A higher rate of aCME eyes had a macula-off retinal detachment (20/24, 83%), compared with nCME eyes (31/70, 44%, P = 0.001). Proliferative vitreoretinopathy (PVR) stage C was more frequent in the aCME group (15/24, 63%) versus the nCME group (5/74, 7%), P < 0.0001. However, both macula-off status (P = 0.06) and presence of PVR C (P = 0.09) lost statistical significance after adjustment for the total number of surgeries performed. Surgical approaches were statistically different between the aCME and nCME groups: primary SB in 1/25 (4%) aCME eyes vs. 25/74 (34%) nCME eyes, PPV in 7/25 (28%) aCME eyes vs. 28/74 (38%) nCME eyes, and combined SB + PPV in 17/25 (68%) aCME eyes vs. 21/74 (28%) nCME eyes (P <0.0001). However, these differences in the surgical approach were not reliably different after adjustment for the number of surgeries. Rates of retinectomy were higher in the aCME group than the nCME group after adjustment (9/25, 36% vs 4/74, 5%, P = 0.009). Rates of cryotherapy were higher in the nCME group (30/74, 41%) than aCME group (4/24, 17%), even after adjustment (P = 0.036). Unadjusted differences in tamponade agent between groups were statistically significant (P < 0.0001). Notably, 16 out of 25 (64%) aCME eyes received silicone oil (SO) at least once, while only 4 out of 74 (5%) of nCME eyes did. However, tamponade differences were no long significant after covariate adjustment. There was no difference in the use of perfluorocarbon liquid (PFCL) (P = 0.728).
At last examination, VA was significantly lower in aCME group (0.85 ± 0.80 LogMAR) than in nCME group (0.20 ± 0.30 LogMAR), P < 0.0001. When adjusting for the number of surgeries and macular detachment, the marginal estimates for between group differences in LogMAR were attenuated (aCME = 0.55 vs nCME = 0.26), though still statistically significant (P = 0.012).
Two patients had non-simultaneous RRDs in each eye. One patient was 23 years of age at the time of both surgeries and underwent SB with cryotherapy in each eye for inferior chronic RRD, without CME development. The other patient was 83 at the time of final surgery in both eyes, had initial surgeries performed elsewhere, had multiple PPVs in both eyes, and received SO in both eyes, and this patient developed cCME in both eye.
OCT characteristics of cCME
Eyes in the cCME group (n = 20) shared particular qualities on OCT (Fig. 2). All eyes had diffuse CME involving the four macular quadrants. The CME always involved the fovea but had variable extent into peripheral macula and was often asymmetric. Cysts were uniformly present in the INL and OPL, with occasional ganglion cell layer involvement. Florid CME often assumed a retinoschitic appearance. With time, cysts coalesced into larger confluent cavities with irregular, polygonal shapes. These cysts often spanned within the same retinal layer and across adjacent layers. Temporary resolution of these cysts after treatment disclosed disorganization and variable atrophy of the retinal layers in areas of cyst confluency. If CME recurred after treatment, it typically recurred in the same distribution of the macula.Fig. 2 Spectral-domain optical coherence tomography and infrared image elevation overlays of two different patients with chronic cystoid macular edema post-rhegmatogenous retinal detachment. The scan in Row A demonstrates schisis-like changes. The scan in Row B demonstrates confluent cystic cavities spanning retinal layers that developed over two years. In both scans, note diffuse, asymmetric distribution of retinal cysts crossing the horizontal raphe, involvement of inner and outer retinal layers, absence of subretinal fluid, and relative preservation of outer retinal bands subjacent to retinal edema
Outer retinal layer integrity was heterogeneous. On the first OCT with CME after the final RRD repair, ellipsoid zone (EZ) disruption was seen in 18 eyes (90%), external limiting membrane (ELM) disruption in 14 eyes (80%), and retinal pigment epithelial (RPE) disruption in 11 eyes (55%). Remarkably, there was no case with subretinal fluid (SRF), and no case of intraretinal hyperreflective foci or hemorrhage.
An epiretinal membrane (ERM) was detectable on OCT during the post-operative follow-up period in 17/20 (85%) cCME eyes, 2/5 (40%) tCME eyes, and 28/74 (38%) of nCME eyes (P = 0.005). Evidence of traction on OCT, such as inner retinal wrinkling or ectopic inner foveal layers, was appreciable in only 4 of the 17 cCME eyes with ERM. However, the severity of CME was out of proportion to the ERM changes in all but one of these four eyes.
CME Treatments
All patients with tCME (n = 5) and cCME (n = 20) received topical medications. Intravitreal injections and surgical interventions were administered according to physician discretion. All patients received corticosteroid drops, non-steroidal anti-inflammatory agent (NSAID) drop, or a combination of both for at least two months after the diagnosis of CME. If the CME failed to respond, patients thereafter received intravitreal injections of anti-VEGF (bevacizumab, ranibizumab, aflibercept), or steroids (triamcinolone acetate (TA), and/or dexamethasone intravitreal implant (DEX)).
The five patients (25%) with tCME had permanent resolution of CME with drops. Table 2 summarizes intravitreal treatments and anatomical responses of cCME. Five patients received at least one bevacizumab (Avastin®, Genentech Inc., San Francisco, CA, USA) injection, and one of these patients also received aflibercept (Eylea®, Regeneron Inc., Tarrytown, NY, USA) injections. In cCME eyes, there was a significant CST (P = 0.016, Wilcoxon signed rank test) and volume (P = 0.016) decrease after TA. (P = 0.125) (Fig. 3). There was no difference in pre-injection CST or volume between groups (P = 0.397, P = 0.457). There was no significant change in CST or volume with anti-VEGF treatment (P =0.915, P = 0.828) or DEX (P = 0.434, P = 0.125). No patient developed elevated intraocular pressure (IOP) after intravitreal injection requiring treatment. One patient developed sterile endophthalmitis after her seventh TA injection that spontaneously resolved without sequelae. A PPV for an ERM was performed in 9/16 cCME eyes with OCT evidence of ERM, with full resolution of the CME in only one eye.Table 2 Treatments for chronic cystoid macular edema (cCME) and anatomical responses on spectral-domain optical coherence tomography
Type of treatment Anti-VEGF TA DEX
Number of eyes 5 7 4
Number of Injections (Median; [Range]) 2.5, 1-14 2.0, 1-10 2.5, 1-7
CST pre-injection (μm) 401 ± 84.9 481 ± 104 397 ± 57.0 P = 0.397
CST post-injection (μm) 393 ± 106 402 ± 102 355 ± 80.4
Percent CST change (μm) − 1.44 ± 17.1, P = 0.915 − 15.6 ± 16.6, P = 0.016 − 11.0 ± 10.7, P = 0.434
Inner macular volume pre-injection (mm3) 2.81 ± 0.43 3.18 ± 0.56 3.12 ± 0.80 P = 0.457
Inner macular volume post-injection (mm3) 2.74 ± 0.53 2.72 ± 0.53 2.66 ± 0.486
Percent (%) inner macular volume change (mm3) −2.49 ± 12.35, P = 0.828 −13.9 ± 10.8, P = 0.016 −10.7 ± 25.7, P = 0.125
Values are listed as averages with standard deviations
VEGF vascular endothelial growth factor, TA triamcinolone acetate, DEX dexamethasone implant, CST central subfield thickness
Fig. 3 Spectral-domain optical coherence tomography (OCT) images of chronic cystoid macular edema (CME) post-rhegmatogenous retinal detachment (RRD) repair of the left eye, with dates and visual acuities (VA). Panel A: OCT prior to dexamethasone implant (DEX) injection. Panel B: OCT 1 month after DEX injection, showing resolution of CME but retinal layer atrophy. Modest VA improvement was noted. Panel C: OCT four months after injection, showing recurrence of CME in a similar distribution and slight decrease in VA
Discussion
Chronic CME after retinal detachment repair remains a challenging complication. In this paper, the risk factors for post-RRD CME, its OCT characteristics, and treatments outcomes are described.
Chronic post-RRD CME is thought to be pathophysiologically distinct from other etiologies of CME [3]. Among CME etiologies such as uveitis, RVO, and DME, many of the cytokines and damaged tissue responses are shared [1, 2, 15]. Certain CME etiologies, however, may have unique pathophysiologic mechanisms despite phenotypic similarities [16]. Entities with a significant pro-angiogenic component, such as exudative AMD, may respond to anti-VEGF agents, while those with a broad inflammatory component, such as uveitic CME or Irvine-Gass syndrome, may respond better to anti-inflammatory drugs [12].
While some studies found no risk factor differences for CME rates [5, 17], some series have, on univariate analyses, reported increased rates in pseudophakic [18] and aphakic eyes [6], older patients, more extensive RRD, and a history of a detached macula. In the present study, lens status was significantly different between groups, with increased pseudophakia and aphakia in aCME eyes. Unicameral communication in vitrectomized eyes modifies circulation of inflammatory cytokines, as animal studies have noted changes in oxygen and antioxidant gradients [19]. Higher rates of pseudophakia/aphakia in the aCME group may be related either to the actual lens surgery or to the complexity of the vitreo-retinal surgeries requiring lens extraction. As a substantial proportion of eyes with complicated RRD will be made pseudophakic or aphakic, anticipating CME in complex cases can have prognostic implications.
Eyes with CME had a greater number of surgeries, higher rates of PVR grade C and retinectomy, and higher rates of SO use. Many studies have shown increased inflammation and CME with more complicated ocular surgeries and inflammatory risk factors [3, 11, 20]. Re-detachments are frequently associated with PVR formation and warrant additional surgeries, both of which can increase intraocular inflammation and possible risk for CME [21]. Retinectomy is helpful when PVR membranes are not amenable to mechanical peeling, and therefore retinectomy likely indicates severe pathology rather than directly causing CME.
Macular detachment was associated with a higher risk of CME, which is in line with prior papers [18]. Of note, previous studies have noted outer nuclear layer CME on OCT of the detached macula [22, 23]. Although the retinal hydration theory, implicated in macular hole edema formation [24], may contribute to post-RRD CME, the presence of leakage on FA suggests dynamic fluid movements as opposed to static, non-leaking cysts. Moreover, absence of SRF after RD repair would theoretically lead to rapid elimination of intraretinal fluid by normal pumping mechanisms. Although such studies for macular detachment and CME development have not been explored [18, 25], permanent damage to retinal cellular elements while detached may lead to persistent dysfunction and contribute to CME.
There was a significant difference in surgical approaches between groups, with higher rates of combined SB and PPV in aCME eyes. This is not surprising, given that scleral buckles are often combined with PPV for complex or recurrent detachments to support the vitreous base and/or areas of retinal pathology. However, there was significantly more cryotherapy in the nCME group. Cryotherapy at our institution is only used during primary scleral buckling, usually for limited and uncomplicated detachments in phakic patients. While data comparing CME rates between PPV and SB are scant, the correlation between more complicated detachments and CME is consistent [3, 11, 18].
After adjusting for the number of surgeries, type of surgery (P = 0.21), macular detachment (P = 0.06), PVR Grade C (P = 0.09) and tamponade type (P = 0.10) lost statistical significance. This may be related to the limited sample size, as there remained a trend towards significance. Moreover, these factors are clinically related to the number of surgeries and surgical failure. The interplay of inflammation among these factors requires more formal study.
Characteristics of CME on OCT can be useful diagnostic clues, and post-RRD cCME displays distinguishing OCT features (Fig. 2). Previous studies have examined OCTs of various conditions associated with CME and noted distinctive findings [13]. These findings could then be used to diagnose conditions accurately as well as account for variability in VA [26]. Post-RRD cCME shares features of uveitic CME, such as diffuse macular distribution, inner and outer layer cysts, and absence of hyperreflective foci. This contrasts to post-RRD tCME, which is much less severe, more central and fleeting, and may be a variant of pseudophakic CME.
The presence of ERM is common after RRD and may confound CME diagnosis [16]. Although there was a significant difference between groups in the presence of ERM on OCT, there was resolution of CME in only one eye after ERM peeling, suggesting that traction plays a small role in most cases of post-RRD CME. Therefore, there should be high suspicion for post-RRD cCME in any patient status-post RRD repair that has severe, diffuse CME without SRF in the absence of other typical inflammatory or tractional signs.
The RPE has a well-studied role in pumping syneretic vitreous fluid through the retina and into the choroidal space [1]. Active fluid transport regulation by the RPE and Muller cells along with maintenance of tight junctional proteins are thought to mitigate CME accumulation [1, 2, 15], and dysfunction of these cells causes an imbalance of fluid inflow and egress. Previous papers examining CME OCT findings note varying SRF rates, from 5% in uveitic CME up to 100% in central RVO-associated CME [1, 4, 13, 27]. Therefore, the absence of SRF in cCME suggests a grossly functioning RPE and outer retinal barrier.
Intravitreal corticosteroids were more effective than intravitreal anti-VEGF or topical medications for cCME in our series. Recent investigations have shown success with intravitreal corticosteroids for chronic post-RRD CME [16, 25]. Thanos et al. found favorable responses to DEX all eyes, but in all cases CME recurred after 3 months. This aligns with pharmacokinetic studies showing a dual-phase response of high dexamethasone concentrations for the first 2 months after delivery followed by a precipitous decrease during the third month [28]. Experimental studies have demonstrated a reduced half-life of anti-VEGF agents and triamcinolone acetate in vitrectomized eyes compared with non-vitrectomized eyes [28, 29], but similar clearances between eyes with DEX. Statistically significance for anatomical improvement was not reached for DEX in our series, likely due to the small number of eyes. Moreover, aphakia has been suggested to cause increased unicameral circulation of inflammatory cytokines [6, 30], but aphakia precludes the use of DEX. One randomized controlled trial evaluating PPV with SO for RRD with grade C PVR found a significant decrease in CME occurrence at 6 months post-operatively in those with intraoperative DEX [31]. Corticosteroids have been shown to modulate a number of cytokines secreted by retinal cells, such as tumor necrosis factor-α, interleukins-1β, 6, and 8, as well as induce expression of occludin, ZO-1, and claudin-5 [1, 16, 31]. Steroids also modulate expression of aquaporin, predominantly expressed in end-feet of Müller cells and astrocytes. Corticosteroids may therefore stabilize the BRB and encourage resolution of CME, accounting for the increased efficacy of corticosteroids over anti-VEGF agents. Nevertheless, disadvantages of TA and DEX include accelerated cataract formation and risk of increased IOP; however, most patients with cCME will require cataract extraction, and no patient in our series required treatment for ocular hypertension.
Average final visit VA was significantly worse in the aCME group even after adjusting for macula-off status and number of surgeries. Reports on recalcitrant CME after PPV for RRD, despite anatomic improvement, found only short-term visual acuity gains [16, 25].
Irvine-Gass syndrome (IGS) is another potential diagnosis in these cases. We did not regularly perform FA or optic disc evaluations to check for optic nerve head leakage during the course of follow up. However, IGS is not described after PPV and has been described as a potential treatment option in many cases [32]. Therefore, IGS would have likely responded to topical treatments, steroid injections, or PPV. The OCT appearance of IGS is also less diffuse, more foveocentric, and may be associated with SRF, as opposed to characteristics noted with post-RRD cCME.
Our paper has a relatively large sample size of post-RRD CME, long-term patient records and follow up, and variety of treatments. Despite this, our study has several limitations. The retrospective analysis precluded standardized imaging and treatment protocols. Significant loss to follow-up likely led to underreporting of chronic post-RRD CME and an inability to accurately determine incidence. The high percentage of CME likely relates to inclusion of eyes that had initial RRD repairs prior to the inclusion period and multiple referrals for complex cases. We were unable to determine after which surgery CME appeared due to inconsistent timing and absence of OCT acquisition between surgeries, or missing outside records. A small number of eyes received anti-VEGF injections, and greater numbers may show CME improvement. A larger, prospective study evaluating complex macular surgeries is warranted.
In conclusion, cCME after RRD is a complex entity with interconnected risk factors. A high index of suspicion based on risk factor and imaging characteristics can allow anticipation of cCME development and early treatment. Currently, corticosteroids have the most evidence of treatment success, and prompt intervention may provide better functional and structural outcomes.
Supplementary information
Additional file 1: Table S1. Descriptive statistics for all three groups, as explained in results population section.
Abbreviations
CMECystoid macular edema
tCMETransient CME
cCMEChronic CME
aCMEAll CME
DRDiabetic retinopathy
AMDAge-related macular degeneration
RVORetinal vein occlusion
RPRetinitis pigmentosa
RRDRhegmatogenous retinal detachment
FAFluorescein angiography
OCTOptical coherence tomography
SBScleral buckle
PPVPars plana vitrectomy
VEGFVascular endothelial growth factor
EHRElectronic health records
CPTCurrent procedural terminology
TATriamcinolone acetonide
DEXDexamethasone intravitreal implant
CSTCentral subfield thickness
ETDRSEarly treatment for diabetic retinopathy study
SDStandard deviation
PVRProliferative vitreoretinopathy
PFCLPerfluorocarbon liquid
INLInner nuclear layer
OPLOuter plexiform layer
EZEllipsoid zone
ELMExternal limiting membrane
ERMEpiretinal membrane
RPERetinal pigment epithelium
SRFSubretinal fluid
NSAIDNon-steroidal anti-inflammatory drug
IOPIntraocular pressure
SOSilicone oil
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Supplementary information accompanies this paper at 10.1186/s40942-020-00254-9.
Acknowledgements
The authors would like to thank Nicholas J. Jackson, PhD, for his statistical assistance and valuable review of the manuscript.
Authors’ contributions
All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by CP, IC, AG, SG, SDS, and JPH. The first draft of the manuscript was written by CP, and all authors commented on draft versions of the manuscript. All authors read and approved the final manuscript.
Funding
The research described was supported by NIH/National Center for Advancing Translational Science (NCATS) UCLA CTSI Grant Number UL1TR001881, and by an unrestricted grant from Research to Prevent Blindness.
Availability of data and materials
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.
Ethics approval and consent to participate
This research study was conducted retrospectively from data obtained for clinical purposes. An IRB official waiver of ethical approval was granted from the IRB of the University of California Los Angeles Office of Human Research Protection (IRB#16-000574).
Consent for publication
Not applicable.
Competing interests
All authors certify that they have no affiliations with or involvement in any organization or entity with any financial interest (such as honoraria; educational grants; participation in speakers’ bureaus; membership, employment, consultancies, stock ownership, or other equity interest; and expert testimony or patent-licensing arrangements), or non-financial interest (such as personal or professional relationships, affiliations, knowledge or beliefs) in the subject matter or materials discussed in this manuscript. | TRIAMCINOLONE ACETONIDE | DrugsGivenReaction | CC BY | 33494835 | 18,924,022 | 2021-01-25 |
What was the outcome of reaction 'Non-infectious endophthalmitis'? | Macular edema after rhegmatogenous retinal detachment repair: risk factors, OCT analysis, and treatment responses.
OBJECTIVE
To investigate risk factors, imaging characteristics, and treatment responses of cystoid macular edema (CME) after rhegmatogenous retinal detachment (RRD) repair.
METHODS
Consecutive, retrospective case-control series of patients who underwent pars plana vitrectomy (PPV) and/or scleral buckling (SB) for RRD, with at least six months of follow-up. Clinical and surgical parameters of patients with and without CME (nCME), based on spectral-domain optical coherence tomography (OCT), were compared.
RESULTS
Of 99 eyes enrolled, 25 had CME while 74 had nCME. Patients with CME underwent greater numbers of surgeries (P < 0.0001). After adjusting for number of surgeries, macula-off RRD (P = 0.06), proliferative vitreoretinopathy (PVR) (P = 0.09), surgical approach (PPV and/or SB, P = 0.21), and tamponade type (P = 0.10) were not statistically significant, although they all achieved significance on univariate analysis (P = 0.001 or less). Intraoperative retinectomy (P = 0.009) and postoperative pseudophakia or aphakia (P = 0.008) were more frequent in the CME group, even after adjustment. Characteristics of cCME on OCT included diffuse distribution, confluent cysts, and absence of subretinal fluid or intraretinal hyperreflective foci. Macular thickness improved significantly with intravitreal triamcinolone (P = 0.016), but not with anti-vascular endothelial growth factor agents (P = 0.828) or dexamethasone implant (P = 0.125). After adjusting for number of surgeries and macular detachment, final visual acuities remained significantly lower in the CME vs nCME group (P = 0.012).
CONCLUSIONS
Risk factors of CME include complex retinal detachment repairs requiring multiple surgeries, and pseudophakic or aphakic lens status. Although this cCME was associated with poor therapeutic response, corticosteroids were the most effective studied treatments.
Background
Cystoid macular edema (CME) is a common retinal condition characterized by macular thickening with intra-retinal fluid accumulation, often accompanied by decreased visual acuity (VA) [1]. It may develop as a complication of a wide spectrum of retinal diseases including diabetic retinopathy (DR), uveitis, exudative age-related macular degeneration (AMD), retinal vein occlusion (RVO), and genetic syndromes such as retinitis pigmentosa (RP) [2].
Although the pathophysiology of CME is multifactorial, breakdown of the inner blood retinal barrier is a common endpoint in most cases [1]. Current theories suggest subclinical inflammation as responsible for post-rhegmatogenous retinal detachment (RRD) CME [3]. While progressive leakage may be outlined with fluorescein angiography (FA) as the gold-standard for CME diagnosis, optical coherence tomography (OCT) is currently the most common imaging modality in the diagnosis and characterization of CME, as it is non-invasive and provides high resolution cross-sectional imaging of retinal anatomy [4], allowing easier and more frequent follow-up,
Rhegmatogenous retinal detachment is characterized by progressive accumulation of subretinal fluid due to retinal breaks. Although surgical repairs, including scleral buckle (SB) and pars plana vitrectomy (PPV), are effective surgical treatments, some cases with successful reattachment may have poor visual outcomes related to postoperative CME development [5, 6], which may persist for years in a minority of patients [7]. Retrospective and observational studies using FA and OCT have shown rates of post-vitrectomy CME varying from 5.5% after PPV for symptomatic floaters to 40% after complicated detachment repairs [6, 8, 9]. Treatments for CME primarily target inflammatory and pro-angiogenic mediators, but standard therapies such as anti-vascular endothelial growth factor (anti-VEGF) therapies may be ineffective for post-RRD CME [9, 10].
There is little data on post-RRD CME risk factors, rates, and anatomical characteristics [3, 5, 11]. Therefore, this observational study was designed to compare a consecutive case series of eyes with versus without post-RRD CME, with the aim to determine its risk factors and describe its clinical characteristics and therapeutic outcomes.
Methods
This was a retrospective, observational study approved by the medical center’s institutional review board, University of California Los Angeles Office of Human Research Protection (IRB#16-000574). This study adhered to the tenets of the Declaration of Helsinki and the rules of the Health Insurance Portability and Accountability Act of 1996.
Electronic health records (EHR) from a large academic referral center (Stein Eye Institute at UCLA) were reviewed. Current Procedural Terminology (CPT) coding records of surgical procedures from January 2015 to December 2017 were queried.
Population
All candidates underwent SB, PPV, or combined procedures for RRD, performed by two experienced vitreoretinal surgeons (JPH and SDS), with at least 6 months of follow up after surgery. Records were evaluated through July 2018.
Exclusion criteria were severe ocular trauma, uveitis, DR, endophthalmitis, RVO, myopic retinoschisis, or advanced dry or wet AMD.
Spectral Domain-OCT Analysis
All patients diagnosed with CME were examined with eye-tracked OCT. All OCTs were acquired with the Spectralis® (Heidelberg Engineering GmbH, Heidelberg, Germany) and RS-3000 (Nidek® Inc, San Jose, CA) devices. All CME was analyzed with Spectralis® OCTs consisting of 19 horizontal B-scans and manually adjusted for foveal centration. All OCT scans were carefully reviewed independently by two graders (CP, JPH) on the Heidelberg Eye Explorer software (Version 1.10.0.0).
A diagnosis of CME was noted if intraretinal hyporeflective spaces were noted in the inner nuclear layer (INL) and/or outer plexiform layer (OPL). Retinal thickness measurements were not used for CME diagnosis, as eyes had varying levels of atrophy.
Eyes were classified as having postoperative transient CME (tCME), chronic CME (cCME), or no CME (nCME). Both tCME and cCME were included as all CME (aCME) for statistical analysis. Postoperative tCME was defined as CME seen on OCT within 6 months of the final RRD, lasting less than 6months, and resolving using topical treatment. Postoperative cCME was defined as CME seen on two OCTs at least 6 months apart, based on previous reports [12].
Recorded characteristics of cCME on OCTs included presence of subretinal fluid, layers of CME involvement, presence of intraretinal hyperreflective foci, and integrity of outer retinal layers. Efficacy of anti-VEGF, triamcinolone acetonide (TA), or dexamethasone implant (Ozurdex®, Allergan Inc, Irvine, California) (DEX) injections were assessed after 4–6 weeks, if OCT was available. To determine treatment effect, pre- and post-injection OCTs were analyzed for central subfield thicknesses (CST) and inner macular volumes, comprised of the central five areas of the standard early treatment for diabetic retinopathy study (ETDRS) subfields [13].
Clinical charts analysis
Preoperative RRD parameters, intraoperative and post-operative data were collected. Glaucoma was counted if the patient carried this diagnose from a glaucoma specialist. Visual acuity was measured on a Snellen chart and converted to logarithm of the minimum angle of resolution (LogMAR) values for statistical analysis. Count fingers and hand motions vision were recorded as 1.98 and 2.28 LogMAR, respectively, based on previous studies using the Freiburg Visual Acuity Test [14]. Type of cCME treatment and number of intravitreal injections were included.
Statistical analysis
Qualitative values were listed as ratios and percentages while quantitative values were presented as mean ± standard deviation (SD). Qualitative variables were compared using the Fisher exact test. To compare continuous data between two groups, a Mann–Whitney U test was used. The Wilcoxon signed rank test was used to analyze changes in CST and inner retinal volume. The Kruskal–Wallis test was used to compare pre-injection OCT parameters between groups. The Shapiro–Wilk test assessed the normality of variable distribution. Covariate adjusted differences between CME groups were assessed using regression modeling (i.e. logistic, linear, and multinomial) using the number of surgeries as the covariate. Final visual acuity (logMAR) was log transformed in multivariable analyses and used the additional covariate of macula on/off. All statistics were performed in Stata SE 15.1 (StataCorp LP, College Station, TX). A P value of less than 0.05 was considered statistically significant. Denominators of ratios were less than the total number of eyes in the category if eyes could not be included in analyses due to missing or incomplete records.
Results
Population
A flowchart of population selection is shown in Fig. 1. A total of 508 surgical records were retrieved using CPT codes from January 2015 to December 2017. Of these, 133 eyes undergoing RRD repair met inclusion and exclusion criteria. Of these, 34 had less than 6 months of follow-up. The remaining 99 eyes of 97 patients were included for analysis. Of these, 20 patients (20%) had cCME, 5 (5%) had tCME, and 74 (75%) had nCME. Our primary analyses examine tCME and cCME as a single group, all CME (aCME), in comparison to nCME due to the small sample size for tCME. Descriptive statistics for all three groups can be found in the Additional file 1: Table S1.Fig. 1 Flowchart of patient selection process. ICD-9: International Classification of Disease, 9th edition. CPT Current Procedural Terminology, CME Cystoid Macular Edema
CME risk factors
Demographic and surgical data are summarized by CME group in Table 1. There was no difference in age at last surgery between patients in the aCME group (64.1 ± 11.6 years) versus patients in the nCME group (56.7 ± 18.0 years, P = 0.092). There was no significant difference in gender (P = 0.093), glaucoma status (P = 0.258), or length of follow-up (P = 0.869). Among those with glaucoma, there was no difference in the rates of topical prostaglandin analogs, other topical medications, or glaucoma surgery between groups (P = 0.992).Table 1 Demographics, baseline characteristics, and surgical data of patients with aCME and nCME
aCME nCME P value Adjusted P value1
Demographic data
Number of eyes 25 (25%) 74 (75%)
Follow-up (months) 21.4 ± 12.1 20.4 ± 10.8 0.87
Sex, Female 8 (32%) 38 (34%) 0.09
Age (years) 64.1 ± 11.6 56.7 ± 18.0 0.09
Clinical data
Right eye 11 (44%) 40 (54%) 0.38
Glaucoma 4 (16%) 6 (8%) 0.26
Lens status < 0.001 0.008
Phakic 1 (4%) 44 (60%)
Pseudophakic 14 (56%) 28 (38%)
Aphakic 10 (40%) 2 (3%)
Macula offa 20/24 (83%) 31/70 (44%) 0.001 0.06
PVR Stage Ca 15/24 (63%) 5/74 (7%) < 0.001 0.09
Final VA (LogMAR) 0.85 ± 0.80 0.20 ± 0.30 < 0.001 0.012b
ERM 18 (72%) 28 (38%) 0.005
Surgical details
Number of surgeries 3.5 ± 1.8 1.4 ± 0.9 < 0.001
Multiple PPV 21 (84%) 17 (23%) < 0.001 –c
Referred after surgery elsewhere 12 (48%) 5 (7%) < 0.001 0.31
Number of surgery outside 1 ± 1.3 0.095 ± 0.4 < 0.001
Type of surgery < 0.001 0.21
SB 1 (4%) 25 (34%)
PPV 7 (28%) 28 (38%)
PPV + SB 17 (68%) 21 (28%)
Tamponade agent < 0.001 0.10
None/Air 1 (4%) 24 (32%)
Gas (SF6 or C3F8) 8 (32%) 46 (62%)
Silicone oil 16 (64%) 4 (5%)
Cryotherapya 4/24 (17%) 30/73 (41%) 0.047 0.036
Retinectomy 9 (36%) 4 (5%) < 0.001 0.009
PFCLa 18/23 (78%) 35/47 (75%) 0.73 0.38
aCME all (chronic + transient) cystoid macular edema, nCME no cystoid macular edema, PVR proliferative vitreoretinopathy, VA visual acuity, LogMAR (logarithm of the minimum angle of resolution), PPV pars-plana vitrectomy, ERM epiretinal membrane, SB scleral buckle, PFCL perfluorocarbon liquid
aDenominators are provided if the number is less than the total number of eyes in the category due to missing or incomplete data
bFinal VA adjusted P value from a model with covariates for total number of surgeries and Macula on/off
cAdjusted model not possible due to collinearity of Multiple PPV with number of surgeries (i.e. those with Multiple PPV had greater than 2 surgeries, while those with no PPV had fewer)
1P-value for difference after adjustment for total number of surgeries
Eyes in the aCME group underwent a significantly greater number of retinal surgeries (3.5 ± 1.8) compared with eyes in the nCME group (1.4 ± 1.9) (P < 0.001). Due to the high collinearity between CME status and number of surgeries, multivariate analysis using this as a covariate was performed. Final lens status differed significantly between groups after adjustment (P = 0.008), with only one eye in the aCME group remaining phakic. A higher rate of aCME eyes had a macula-off retinal detachment (20/24, 83%), compared with nCME eyes (31/70, 44%, P = 0.001). Proliferative vitreoretinopathy (PVR) stage C was more frequent in the aCME group (15/24, 63%) versus the nCME group (5/74, 7%), P < 0.0001. However, both macula-off status (P = 0.06) and presence of PVR C (P = 0.09) lost statistical significance after adjustment for the total number of surgeries performed. Surgical approaches were statistically different between the aCME and nCME groups: primary SB in 1/25 (4%) aCME eyes vs. 25/74 (34%) nCME eyes, PPV in 7/25 (28%) aCME eyes vs. 28/74 (38%) nCME eyes, and combined SB + PPV in 17/25 (68%) aCME eyes vs. 21/74 (28%) nCME eyes (P <0.0001). However, these differences in the surgical approach were not reliably different after adjustment for the number of surgeries. Rates of retinectomy were higher in the aCME group than the nCME group after adjustment (9/25, 36% vs 4/74, 5%, P = 0.009). Rates of cryotherapy were higher in the nCME group (30/74, 41%) than aCME group (4/24, 17%), even after adjustment (P = 0.036). Unadjusted differences in tamponade agent between groups were statistically significant (P < 0.0001). Notably, 16 out of 25 (64%) aCME eyes received silicone oil (SO) at least once, while only 4 out of 74 (5%) of nCME eyes did. However, tamponade differences were no long significant after covariate adjustment. There was no difference in the use of perfluorocarbon liquid (PFCL) (P = 0.728).
At last examination, VA was significantly lower in aCME group (0.85 ± 0.80 LogMAR) than in nCME group (0.20 ± 0.30 LogMAR), P < 0.0001. When adjusting for the number of surgeries and macular detachment, the marginal estimates for between group differences in LogMAR were attenuated (aCME = 0.55 vs nCME = 0.26), though still statistically significant (P = 0.012).
Two patients had non-simultaneous RRDs in each eye. One patient was 23 years of age at the time of both surgeries and underwent SB with cryotherapy in each eye for inferior chronic RRD, without CME development. The other patient was 83 at the time of final surgery in both eyes, had initial surgeries performed elsewhere, had multiple PPVs in both eyes, and received SO in both eyes, and this patient developed cCME in both eye.
OCT characteristics of cCME
Eyes in the cCME group (n = 20) shared particular qualities on OCT (Fig. 2). All eyes had diffuse CME involving the four macular quadrants. The CME always involved the fovea but had variable extent into peripheral macula and was often asymmetric. Cysts were uniformly present in the INL and OPL, with occasional ganglion cell layer involvement. Florid CME often assumed a retinoschitic appearance. With time, cysts coalesced into larger confluent cavities with irregular, polygonal shapes. These cysts often spanned within the same retinal layer and across adjacent layers. Temporary resolution of these cysts after treatment disclosed disorganization and variable atrophy of the retinal layers in areas of cyst confluency. If CME recurred after treatment, it typically recurred in the same distribution of the macula.Fig. 2 Spectral-domain optical coherence tomography and infrared image elevation overlays of two different patients with chronic cystoid macular edema post-rhegmatogenous retinal detachment. The scan in Row A demonstrates schisis-like changes. The scan in Row B demonstrates confluent cystic cavities spanning retinal layers that developed over two years. In both scans, note diffuse, asymmetric distribution of retinal cysts crossing the horizontal raphe, involvement of inner and outer retinal layers, absence of subretinal fluid, and relative preservation of outer retinal bands subjacent to retinal edema
Outer retinal layer integrity was heterogeneous. On the first OCT with CME after the final RRD repair, ellipsoid zone (EZ) disruption was seen in 18 eyes (90%), external limiting membrane (ELM) disruption in 14 eyes (80%), and retinal pigment epithelial (RPE) disruption in 11 eyes (55%). Remarkably, there was no case with subretinal fluid (SRF), and no case of intraretinal hyperreflective foci or hemorrhage.
An epiretinal membrane (ERM) was detectable on OCT during the post-operative follow-up period in 17/20 (85%) cCME eyes, 2/5 (40%) tCME eyes, and 28/74 (38%) of nCME eyes (P = 0.005). Evidence of traction on OCT, such as inner retinal wrinkling or ectopic inner foveal layers, was appreciable in only 4 of the 17 cCME eyes with ERM. However, the severity of CME was out of proportion to the ERM changes in all but one of these four eyes.
CME Treatments
All patients with tCME (n = 5) and cCME (n = 20) received topical medications. Intravitreal injections and surgical interventions were administered according to physician discretion. All patients received corticosteroid drops, non-steroidal anti-inflammatory agent (NSAID) drop, or a combination of both for at least two months after the diagnosis of CME. If the CME failed to respond, patients thereafter received intravitreal injections of anti-VEGF (bevacizumab, ranibizumab, aflibercept), or steroids (triamcinolone acetate (TA), and/or dexamethasone intravitreal implant (DEX)).
The five patients (25%) with tCME had permanent resolution of CME with drops. Table 2 summarizes intravitreal treatments and anatomical responses of cCME. Five patients received at least one bevacizumab (Avastin®, Genentech Inc., San Francisco, CA, USA) injection, and one of these patients also received aflibercept (Eylea®, Regeneron Inc., Tarrytown, NY, USA) injections. In cCME eyes, there was a significant CST (P = 0.016, Wilcoxon signed rank test) and volume (P = 0.016) decrease after TA. (P = 0.125) (Fig. 3). There was no difference in pre-injection CST or volume between groups (P = 0.397, P = 0.457). There was no significant change in CST or volume with anti-VEGF treatment (P =0.915, P = 0.828) or DEX (P = 0.434, P = 0.125). No patient developed elevated intraocular pressure (IOP) after intravitreal injection requiring treatment. One patient developed sterile endophthalmitis after her seventh TA injection that spontaneously resolved without sequelae. A PPV for an ERM was performed in 9/16 cCME eyes with OCT evidence of ERM, with full resolution of the CME in only one eye.Table 2 Treatments for chronic cystoid macular edema (cCME) and anatomical responses on spectral-domain optical coherence tomography
Type of treatment Anti-VEGF TA DEX
Number of eyes 5 7 4
Number of Injections (Median; [Range]) 2.5, 1-14 2.0, 1-10 2.5, 1-7
CST pre-injection (μm) 401 ± 84.9 481 ± 104 397 ± 57.0 P = 0.397
CST post-injection (μm) 393 ± 106 402 ± 102 355 ± 80.4
Percent CST change (μm) − 1.44 ± 17.1, P = 0.915 − 15.6 ± 16.6, P = 0.016 − 11.0 ± 10.7, P = 0.434
Inner macular volume pre-injection (mm3) 2.81 ± 0.43 3.18 ± 0.56 3.12 ± 0.80 P = 0.457
Inner macular volume post-injection (mm3) 2.74 ± 0.53 2.72 ± 0.53 2.66 ± 0.486
Percent (%) inner macular volume change (mm3) −2.49 ± 12.35, P = 0.828 −13.9 ± 10.8, P = 0.016 −10.7 ± 25.7, P = 0.125
Values are listed as averages with standard deviations
VEGF vascular endothelial growth factor, TA triamcinolone acetate, DEX dexamethasone implant, CST central subfield thickness
Fig. 3 Spectral-domain optical coherence tomography (OCT) images of chronic cystoid macular edema (CME) post-rhegmatogenous retinal detachment (RRD) repair of the left eye, with dates and visual acuities (VA). Panel A: OCT prior to dexamethasone implant (DEX) injection. Panel B: OCT 1 month after DEX injection, showing resolution of CME but retinal layer atrophy. Modest VA improvement was noted. Panel C: OCT four months after injection, showing recurrence of CME in a similar distribution and slight decrease in VA
Discussion
Chronic CME after retinal detachment repair remains a challenging complication. In this paper, the risk factors for post-RRD CME, its OCT characteristics, and treatments outcomes are described.
Chronic post-RRD CME is thought to be pathophysiologically distinct from other etiologies of CME [3]. Among CME etiologies such as uveitis, RVO, and DME, many of the cytokines and damaged tissue responses are shared [1, 2, 15]. Certain CME etiologies, however, may have unique pathophysiologic mechanisms despite phenotypic similarities [16]. Entities with a significant pro-angiogenic component, such as exudative AMD, may respond to anti-VEGF agents, while those with a broad inflammatory component, such as uveitic CME or Irvine-Gass syndrome, may respond better to anti-inflammatory drugs [12].
While some studies found no risk factor differences for CME rates [5, 17], some series have, on univariate analyses, reported increased rates in pseudophakic [18] and aphakic eyes [6], older patients, more extensive RRD, and a history of a detached macula. In the present study, lens status was significantly different between groups, with increased pseudophakia and aphakia in aCME eyes. Unicameral communication in vitrectomized eyes modifies circulation of inflammatory cytokines, as animal studies have noted changes in oxygen and antioxidant gradients [19]. Higher rates of pseudophakia/aphakia in the aCME group may be related either to the actual lens surgery or to the complexity of the vitreo-retinal surgeries requiring lens extraction. As a substantial proportion of eyes with complicated RRD will be made pseudophakic or aphakic, anticipating CME in complex cases can have prognostic implications.
Eyes with CME had a greater number of surgeries, higher rates of PVR grade C and retinectomy, and higher rates of SO use. Many studies have shown increased inflammation and CME with more complicated ocular surgeries and inflammatory risk factors [3, 11, 20]. Re-detachments are frequently associated with PVR formation and warrant additional surgeries, both of which can increase intraocular inflammation and possible risk for CME [21]. Retinectomy is helpful when PVR membranes are not amenable to mechanical peeling, and therefore retinectomy likely indicates severe pathology rather than directly causing CME.
Macular detachment was associated with a higher risk of CME, which is in line with prior papers [18]. Of note, previous studies have noted outer nuclear layer CME on OCT of the detached macula [22, 23]. Although the retinal hydration theory, implicated in macular hole edema formation [24], may contribute to post-RRD CME, the presence of leakage on FA suggests dynamic fluid movements as opposed to static, non-leaking cysts. Moreover, absence of SRF after RD repair would theoretically lead to rapid elimination of intraretinal fluid by normal pumping mechanisms. Although such studies for macular detachment and CME development have not been explored [18, 25], permanent damage to retinal cellular elements while detached may lead to persistent dysfunction and contribute to CME.
There was a significant difference in surgical approaches between groups, with higher rates of combined SB and PPV in aCME eyes. This is not surprising, given that scleral buckles are often combined with PPV for complex or recurrent detachments to support the vitreous base and/or areas of retinal pathology. However, there was significantly more cryotherapy in the nCME group. Cryotherapy at our institution is only used during primary scleral buckling, usually for limited and uncomplicated detachments in phakic patients. While data comparing CME rates between PPV and SB are scant, the correlation between more complicated detachments and CME is consistent [3, 11, 18].
After adjusting for the number of surgeries, type of surgery (P = 0.21), macular detachment (P = 0.06), PVR Grade C (P = 0.09) and tamponade type (P = 0.10) lost statistical significance. This may be related to the limited sample size, as there remained a trend towards significance. Moreover, these factors are clinically related to the number of surgeries and surgical failure. The interplay of inflammation among these factors requires more formal study.
Characteristics of CME on OCT can be useful diagnostic clues, and post-RRD cCME displays distinguishing OCT features (Fig. 2). Previous studies have examined OCTs of various conditions associated with CME and noted distinctive findings [13]. These findings could then be used to diagnose conditions accurately as well as account for variability in VA [26]. Post-RRD cCME shares features of uveitic CME, such as diffuse macular distribution, inner and outer layer cysts, and absence of hyperreflective foci. This contrasts to post-RRD tCME, which is much less severe, more central and fleeting, and may be a variant of pseudophakic CME.
The presence of ERM is common after RRD and may confound CME diagnosis [16]. Although there was a significant difference between groups in the presence of ERM on OCT, there was resolution of CME in only one eye after ERM peeling, suggesting that traction plays a small role in most cases of post-RRD CME. Therefore, there should be high suspicion for post-RRD cCME in any patient status-post RRD repair that has severe, diffuse CME without SRF in the absence of other typical inflammatory or tractional signs.
The RPE has a well-studied role in pumping syneretic vitreous fluid through the retina and into the choroidal space [1]. Active fluid transport regulation by the RPE and Muller cells along with maintenance of tight junctional proteins are thought to mitigate CME accumulation [1, 2, 15], and dysfunction of these cells causes an imbalance of fluid inflow and egress. Previous papers examining CME OCT findings note varying SRF rates, from 5% in uveitic CME up to 100% in central RVO-associated CME [1, 4, 13, 27]. Therefore, the absence of SRF in cCME suggests a grossly functioning RPE and outer retinal barrier.
Intravitreal corticosteroids were more effective than intravitreal anti-VEGF or topical medications for cCME in our series. Recent investigations have shown success with intravitreal corticosteroids for chronic post-RRD CME [16, 25]. Thanos et al. found favorable responses to DEX all eyes, but in all cases CME recurred after 3 months. This aligns with pharmacokinetic studies showing a dual-phase response of high dexamethasone concentrations for the first 2 months after delivery followed by a precipitous decrease during the third month [28]. Experimental studies have demonstrated a reduced half-life of anti-VEGF agents and triamcinolone acetate in vitrectomized eyes compared with non-vitrectomized eyes [28, 29], but similar clearances between eyes with DEX. Statistically significance for anatomical improvement was not reached for DEX in our series, likely due to the small number of eyes. Moreover, aphakia has been suggested to cause increased unicameral circulation of inflammatory cytokines [6, 30], but aphakia precludes the use of DEX. One randomized controlled trial evaluating PPV with SO for RRD with grade C PVR found a significant decrease in CME occurrence at 6 months post-operatively in those with intraoperative DEX [31]. Corticosteroids have been shown to modulate a number of cytokines secreted by retinal cells, such as tumor necrosis factor-α, interleukins-1β, 6, and 8, as well as induce expression of occludin, ZO-1, and claudin-5 [1, 16, 31]. Steroids also modulate expression of aquaporin, predominantly expressed in end-feet of Müller cells and astrocytes. Corticosteroids may therefore stabilize the BRB and encourage resolution of CME, accounting for the increased efficacy of corticosteroids over anti-VEGF agents. Nevertheless, disadvantages of TA and DEX include accelerated cataract formation and risk of increased IOP; however, most patients with cCME will require cataract extraction, and no patient in our series required treatment for ocular hypertension.
Average final visit VA was significantly worse in the aCME group even after adjusting for macula-off status and number of surgeries. Reports on recalcitrant CME after PPV for RRD, despite anatomic improvement, found only short-term visual acuity gains [16, 25].
Irvine-Gass syndrome (IGS) is another potential diagnosis in these cases. We did not regularly perform FA or optic disc evaluations to check for optic nerve head leakage during the course of follow up. However, IGS is not described after PPV and has been described as a potential treatment option in many cases [32]. Therefore, IGS would have likely responded to topical treatments, steroid injections, or PPV. The OCT appearance of IGS is also less diffuse, more foveocentric, and may be associated with SRF, as opposed to characteristics noted with post-RRD cCME.
Our paper has a relatively large sample size of post-RRD CME, long-term patient records and follow up, and variety of treatments. Despite this, our study has several limitations. The retrospective analysis precluded standardized imaging and treatment protocols. Significant loss to follow-up likely led to underreporting of chronic post-RRD CME and an inability to accurately determine incidence. The high percentage of CME likely relates to inclusion of eyes that had initial RRD repairs prior to the inclusion period and multiple referrals for complex cases. We were unable to determine after which surgery CME appeared due to inconsistent timing and absence of OCT acquisition between surgeries, or missing outside records. A small number of eyes received anti-VEGF injections, and greater numbers may show CME improvement. A larger, prospective study evaluating complex macular surgeries is warranted.
In conclusion, cCME after RRD is a complex entity with interconnected risk factors. A high index of suspicion based on risk factor and imaging characteristics can allow anticipation of cCME development and early treatment. Currently, corticosteroids have the most evidence of treatment success, and prompt intervention may provide better functional and structural outcomes.
Supplementary information
Additional file 1: Table S1. Descriptive statistics for all three groups, as explained in results population section.
Abbreviations
CMECystoid macular edema
tCMETransient CME
cCMEChronic CME
aCMEAll CME
DRDiabetic retinopathy
AMDAge-related macular degeneration
RVORetinal vein occlusion
RPRetinitis pigmentosa
RRDRhegmatogenous retinal detachment
FAFluorescein angiography
OCTOptical coherence tomography
SBScleral buckle
PPVPars plana vitrectomy
VEGFVascular endothelial growth factor
EHRElectronic health records
CPTCurrent procedural terminology
TATriamcinolone acetonide
DEXDexamethasone intravitreal implant
CSTCentral subfield thickness
ETDRSEarly treatment for diabetic retinopathy study
SDStandard deviation
PVRProliferative vitreoretinopathy
PFCLPerfluorocarbon liquid
INLInner nuclear layer
OPLOuter plexiform layer
EZEllipsoid zone
ELMExternal limiting membrane
ERMEpiretinal membrane
RPERetinal pigment epithelium
SRFSubretinal fluid
NSAIDNon-steroidal anti-inflammatory drug
IOPIntraocular pressure
SOSilicone oil
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Supplementary information accompanies this paper at 10.1186/s40942-020-00254-9.
Acknowledgements
The authors would like to thank Nicholas J. Jackson, PhD, for his statistical assistance and valuable review of the manuscript.
Authors’ contributions
All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by CP, IC, AG, SG, SDS, and JPH. The first draft of the manuscript was written by CP, and all authors commented on draft versions of the manuscript. All authors read and approved the final manuscript.
Funding
The research described was supported by NIH/National Center for Advancing Translational Science (NCATS) UCLA CTSI Grant Number UL1TR001881, and by an unrestricted grant from Research to Prevent Blindness.
Availability of data and materials
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.
Ethics approval and consent to participate
This research study was conducted retrospectively from data obtained for clinical purposes. An IRB official waiver of ethical approval was granted from the IRB of the University of California Los Angeles Office of Human Research Protection (IRB#16-000574).
Consent for publication
Not applicable.
Competing interests
All authors certify that they have no affiliations with or involvement in any organization or entity with any financial interest (such as honoraria; educational grants; participation in speakers’ bureaus; membership, employment, consultancies, stock ownership, or other equity interest; and expert testimony or patent-licensing arrangements), or non-financial interest (such as personal or professional relationships, affiliations, knowledge or beliefs) in the subject matter or materials discussed in this manuscript. | Recovered | ReactionOutcome | CC BY | 33494835 | 18,924,022 | 2021-01-25 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Colorectal cancer metastatic'. | Whole genome sequencing of metastatic colorectal cancer reveals prior treatment effects and specific metastasis features.
In contrast to primary colorectal cancer (CRC) little is known about the genomic landscape of metastasized CRC. Here we present whole genome sequencing data of metastases of 429 CRC patients participating in the pan-cancer CPCT-02 study (NCT01855477). Unsupervised clustering using mutational signature patterns highlights three major patient groups characterized by signatures known from primary CRC, signatures associated with received prior treatments, and metastasis-specific signatures. Compared to primary CRC, we identify additional putative (non-coding) driver genes and increased frequencies in driver gene mutations. In addition, we identify specific genes preferentially affected by microsatellite instability. CRC-specific 1kb-10Mb deletions, enriched for common fragile sites, and LINC00672 mutations are associated with response to treatment in general, whereas FBXW7 mutations predict poor response specifically to EGFR-targeted treatment. In conclusion, the genomic landscape of mCRC shows defined changes compared to primary CRC, is affected by prior treatments and contains features with potential clinical relevance.
Introduction
Primary colorectal cancer (CRC) can be divided into a major group of chromosomally instable tumors and a minor group of hypermutated, chromosomally stable tumors due to microsatellite instability (MSI) or POLE mutations1. Parallel to the described genomic subtype division, transcriptomic analysis was used to identify four consensus molecular subtypes (CMSs) with distinguishing features including prognosis2.
Molecular analysis of CRC revealed specific genetic alterations with clinical implications. Mutations in KRAS and BRAF predict failure to treatment with EGFR-inhibitors, whereas copy number alterations of ERBB2 or IGF2, and the occurrence of chromosomal translocations leading to fusion genes such as NAV2/TCF7L1, are potentially drug targetable1,3.
Although the molecular knowledge of primary CRC has contributed to a better understanding of its pathogenesis, cancer-related mortality usually occurs as a consequence of distant metastases, in which ongoing mutational processes and selective treatment pressure can result in altered molecular characteristics4.
To date, in-depth analyses of large series of colorectal cancer metastases are limited to studies using either whole-exome sequencing (WES) or targeted sequencing of cancer-associated genes4–6. Although these studies yielded extensive knowledge on the presence of specific genomic aberrations in mCRC, they do not necessarily reflect its complete molecular landscape. For optimal identification of mutational signatures, the power provided by whole-genome sequencing (WGS) data greatly exceeds that of WES7. Next to this, WGS simultaneously allows for the determination of MSI, structural rearrangements, chromothripsis, and kataegis. In addition, clinically relevant genetic alterations within noncoding regions were recently reported in primary CRC8. To date, the only other study which reported in detail on WGS data of colorectal metastases included 12 patients4.
Here, we provide a comprehensive description of the molecular landscape of metastatic CRC (mCRC). We use WGS data obtained from a large multicenter, prospective collection of snap-frozen metastatic tissue biopsies from 429 patients starting a new line of systemic treatment9. In addition, matched RNA-seq data are available for 91 patients. The observed metastatic molecular landscape is compared to WGS data of primary CRC cohorts (Supplementary Table 1), associated with prior treatments as well as treatment response, and evaluated for clinical utility.
Results
Cohort description
Clinical characteristics of our included cohort of 429 patients are summarized in Table 1. Median tumor purity (0.53 (IQR 0.38–0.67) was estimated on the obtained sequencing data and was not significantly different between biopsy sites. Based on a previously described WGS data analysis algorithm9 14 samples (3%) were scored as microsatellite instable (MSI), which is in concordance with the observed MSI frequency in mCRC in literature (4%)10.Table 1 Cohort description.
Patient details Number of patients
Total cohort 429
Gender Female 174
Male 255
Age (median (IQR,Range)) 64 (IQR 56–72, range 25–88)
Prior-treatment details
Systemic prior treatment: yes 284
Treatment regimen Manuscript code
5-FU/capecitabine–oxaliplatin doublet (CAPOX, FOLFOX) PLAT/PYR 121
+bevacizumab PLAT/PYR + targeted 134
5-FU—Topisomerase inhibitor doublet (Irinotecan based, FOLFIRI) TOP/PYR 26
+bevacizumab/panitumumab TOP/PYR + targeted 9
5-FU/capecitabine monotherapy PYRmon 39
+bevacizumab PYR + targeted 36
Topoisomerase inhibitor (Irinotecan) monotherapy TOPmono 67
+bevacizumab TOP + targeted 7
Oxaliplatin + bevacizumab/panitumumab PLAT + targeted 5
Panitumumab/cetuximab/encorafenib+binimetinib/bevacizumab/regorafenib Targeted mono 35
5-FU/capecitabine–oxaliplatin–irinotecan triplet (FOLFOXIRI) CHEMCOM 2
+bevacizumab CHEMCOM + targeted 5
Other Diverse 15
Systemic prior treatment: no 124
Systemic prior treatment: unknown 21
Radiotherapeutic prior treatment: yes 109
RT + systemic treatment 68
Chemoradiation 33
Radiotherapy only 8
Radiotherapeutic prior treatment: no 299
Radiotherapeutic prior treatment: unknown 21
Biopsy details Number of patients
Origin
Liver 287
Soft tissue 84
Lymph node 24
Lung 21
Other 13
Technical details
Tumor purity 0.53 (IQR 0.38–0.67)
Read coverage (median) 102.6× (IQR 94.6×–112.0×)
Patient Characteristics. PLAT/PYR (5-FU/capecitabine–oxaliplatin doublet (CAPOX, FOLFOX)), PLAT/PYR + targeted (bevacizumab added), TOP/PYR + targeted (bevacizumab added), PYRmono (5-FU/capecitabine monotherapy), PYR + targeted (bevacizumab added), TOP/PYR (5-FU-Topisomerase inhibitor doublet (Irinotecan based)), TOPmono (Topoisomerase inhibitor (Irinotecan) monotherapy), TOP + targeted (bevacizumab added), PLAT + targeted (Oxaliplatin + bevacizumab/panitumumab), Targeted mono (panitumumab, cetuximab, encorafenib + binimetinib, bevacizumab, regorafenib), CHEMCOM (5-FU/capecitabine–oxaliplatin–irinotecan triplet (FOLFOXIRI)), CHEMCOM + targeted (bevacizumab added), Other (diverse).
Based on the treatment data, the cohort can be divided in patients who did (n = 284) and who did not (n = 124) receive any systemic treatment prior to the moment the biopsy was taken. Within the group of prior-treated patients, 13 different combinations of treatment regimens were defined as specified in the materials and methods and listed in Table 1.
For 91 cases RNA-seq data were available, allowing us to determine their Consensus Molecular Subtype (CMS). Remarkably, using the CMS-classifier package, none of the metastatic CRC samples were classified as CMS3, whereas 10 were classified as CMS1, 41 as CMS2, and 14 as CMS4. The remaining 26 samples (29%) could not be classified into one of the 4 subtypes, which might be partly due to the presence of normal cells of noncolon origin in our metastatic setting. Indeed, using the alternative CMSCaller algorithm, which is less dependent on signals from the tumor microenvironment, reduced the number of unclassified samples to 14 (15%), whereas still only 3 samples were classified as CMS311. Twenty-two samples were classified as CMS1, 25 as CMS2, 3 as CMS3, and 27 as CMS4.
Regardless of the calling algorithm used, the estimated tumor cell percentage was significantly lower in biopsies classified as CMS4 than in the other subtypes (medians CMS1: 52.5 and 45%; CMS2 61 and 61%; CMS3: none and 66% and CMS4: 34.5 and 42%; KWH; p = 0.0007 and p = 0.0156 for CMS Classifier and CMSCaller, respectively), which is concordant with the described high-stroma content in this subtype2.
The molecular landscape of mCRC
From the WGS data of all 429 cases, we distilled somatically acquired single nucleotide variants (SNVs), multiple nucleotide variants (MNVs), structural variants (SVs), insertions/deletions (InDels), and copy number variants (CNVs). The overall tumor mutational burden (TMB) representing the amount of SNVs, MNVs and InDels per Megabase (Mb), ranged from 0.96 to 366.15 with a median of 7.01 (95% CI 6.62–7.47). Using GISTIC2.0, we identified 55 recurrent CNVs (29 gains and 26 losses) within our entire cohort, containing a number of already known and putative driver genes (Supplementary Data 1). Chromothripsis was observed in 47 cases (11%), whereas kataegis was observed in 102 cases (24%), involving just a single chromosomal region in two-third of cases, with a maximum of 10 regions in one single case. Presence of kataegis was associated with MSI and high TMB (≥10; test for trend p = 0.00014). In fact, 9 out of 13 MSI cases had at least two kataegis regions.
We further evaluated the type and size of SVs observed in our cohort (Fig. 1). A broad range of differently sized Tandem Duplications (TD; ~14–93 kb) with a peak at 26 kb was observed, which was clearly distinct from the TD sizes previously observed in other cancers (~11 kb in BRCA1-mutated, ~231 kb in CCNE1-activated, and ~1.7 Mb TDs in CDK12-mutated cancer, respectively)12. Inversions in mCRC are usually over 10 Mb in size, while deletions range from ~10 kb to 1 Mb, with a distinct peak at ~128 kb. Events within this latter peak include many recurrent deletions in known Common Fragile Site (CFS) genes: e.g., FHIT, RBFOX1, and MACROD2. This phenomenon involving frequent deletions of CSF genes was recently described in primary CRC as well13.Fig. 1 Size distributions of the different types of structural variants.
Ridge-plot of the density of genomic sizes of structural variants in metastatic CRC. INV inversions (blue), DUP tandem duplications (purple), DEL deletions (orange). Source data are provided as a Source Data file.
Using the ratio of nonsynonymous to synonymous substitutions caused by the somatic nucleotide mutations (SNV and InDels; dN/dS analysis), 23 genes were identified as putative driver genes (q < 0.05, Fig. 2, Table 2). In 99.1% of cases (425 out of 429) at least one of these 23 putative driver genes was mutated. Testing for mutual exclusivity only revealed already known associations: KRAS with BRAF/NRAS/RNF43/TP53 (q = 1.06E-7, q = 1.54E-4, q = 0.004, and q = 0.017, respectively), and APC with RNF43/BRAF (both q = 1.54E-4; Supplementary Fig. 1). For those genes also present in the targeted panel used by Yaeger et al.6, comparable mutation frequencies were observed in both cohorts (Table 2).Fig. 2 Oncoplot of metastatic CRC depicting identified driver genes and somatic mutations (SNV, InDels, and MNV).
Top panel: genes identified by dN/dS as driver genes per type of mutation; purple: frameshift variant; orange: other variant; blue: stop/gain variant; green: structural variant. Bottom panel: first track: clinical information: sex (male: orange; female: green) and second track: biopsy site. Track three (PLAT/PYR ± targeted) indicates which patients have been treated with platinum-based therapy (PLAT; e.g., oxaliplatin) and a pyrimidine-targeting drug (PYR; e.g., 5-FU), with or without the addition of another targeted treatment (±targeted; e.g., bevacizumab). Tracks four to six depict the distribution of the consensus molecular subtypes (CMS), tumor mutational burden (TMB), and the number of structural variant deletions of size 10kb–1Mb (DEL_CFS), partly associated with Common Fragile Sites (CFS), respectively. Source data are provided as a Source Data file.
Table 2 Mutation frequency driver genes.
Metastatic CRC Primary CRC—(TCGA, combined studies cBioportal) Metastatic CRC—(Yaeger et al.)
Gene dN/dS q-value Mutations (N) Mutations (%) Mutations (N) Mutations (%) Fisher p-value FDR Hochberg % change in meta Mutations (N) Mutations (%) Fisher p-value FDR Hochberg
TP53 0 317 73.9 1123 57.6 2.04E-10 4.7E-09 16.3 246 76.6 0.395 1
ZFP36L2 0 42 9.8 97 5.0 3.61E-04 0.008 4.8 Not present
KRAS 0 203 47.3 744 38.2 5.88E-04 0.012 9.1 127 39.6 0.037 0.487
APC 0 336 78.3 1372 70.4 8.86E-04 0.018 7.9 241 75.1 0.335 1
PIK3CA 0 68 15.9 445 22.8 0.001 0.023 −7.0 49 15.3 0.840 1
B2M 5.37E-03 8 1.9 91 4.7 0.007 0.128 −2.8 2 0.6 0.202 1
SMAD4 0 74 17.2 243 12.5 0.010 0.164 4.8 47 14.6 0.367 1
ATM 9.50E-04 33 7.7 227 11.6 0.017 0.266 −4.0 18 5.6 0.306 1
FBXW7 0 51 11.9 301 15.4 0.061 0.912 −3.6 25 7.8 0.068 0.814
AMER1 0 37 8.6 209 10.7 0.220 0.913 −2.1 11 3.4 0.004 0.067
ARID1A 1.13E-09 39 9.1 201 10.3 0.480 0.913 −1.2 15 4.7 0.022 0.333
BCL9 5.27E-02 28 6.5 107 5.5 0.419 0.913 1.0 Not present
BCL9L 2.24E-06 27 6.3 133 6.8 0.750 0.913 −0.5 Not present
BRAF 0 56 13.1 273 14.0 0.644 0.913 −1.0 38 11.8 0.657 1
ELF3 5.37E-03 7 1.6 51 2.6 0.299 0.913 −1.0 Not present
LMTK3 1.33E-02 15 3.5 56 2.9 0.530 0.913 0.6 not present
NRAS 0 26 6.1 125 6.4 0.913 0.913 −0.4 14 4.4 0.329 1
PTEN 4.30E-08 17 4.0 123 6.3 0.069 0.913 −2.3 14 4.4 0.854 1
RNF43 2.20E-02 28 6.5 162 8.3 0.239 0.913 −1.8 21 6.5 1.000 1
SMAD3 5.68E-03 11 2.6 72 3.7 0.309 0.913 −1.1 11 3.4 0.518 1
SOX9 0 41 9.6 177 9.1 0.782 0.913 0.5 16 5.0 0.025 0.352
TCF7L2 1.07E-09 50 11.7 177 9.1 0.103 0.913 2.6 19 5.9 0.007 0.116
TGIF1 1.33E-02 18 4.2 62 3.2 0.300 0.913 1.0 Not present
Twenty-three genes identified as putative driver genes using the ratio of nonsynonymous to synonymous substitutions caused by the somatic nucleotide mutations (SNV and InDels; dN/dS analysis). P-values are derived from the Fisher exact test (two-sided) and corrected for multiple testing using the FDR (Hochberg) method.
Similarly, for 15 noncoding genes an enriched mutation rate was observed compared to surrounding nonannotated regions (Table 3), suggesting these genes are relevant for the oncogenic process. These noncoding genes include PTENP1, a known tumor suppressor in CRC14, MALAT1, for which an increased mutation rate was already described in a pan-cancer analysis15, and LINC00672, described to promote chemo-sensitivity16.Table 3 Mutation frequency noncoding genes.
Metastatic CRC Primary CRC - (ICGC)
ENSG Symbol Size Chr Type Mutation rate FDR Hochberg Mutations (N) Mutations (%) Mutations (N) Mutations (%) Fisher p-value FDR Hochberg
ENSG00000273001 AL731533.2 577 10 lncRNA 0.067591 0 6 1.4 0 0 0.001291 3.87E-03
ENSG00000280325 AC074183.2 921 7 TEC 0.033659 0 25 5.8 a a
ENSG00000261584 AL513548.1 1723 6 lncRNA 0.012768 9.93E-24 14 3.3 3 0.3 3.84E-05 2.09E-04
ENSG00000259834 AL365361.1 3480 1 lncRNA 0.007184 2.48E-09 17 4.0 0 0 5.62E-09 7.30E-08
ENSG00000264920 AC018521.5 4583 17 lncRNA 0.00851 1.02E-08 16 3.7 1 0.1 2.06E-07 1.85E-06
ENSG00000231784 DBIL5P 2676 17 Transcribed_unitary_pseudogene 0.008595 6.26E-07 18 4.2 1 0.1 2.39E-08 2.63E-07
ENSG00000273033 LINC02035 5475 3 lncRNA 0.006758 9.54E-06 23 5.4 0 0.0 6.15E-12 8.61E-11
ENSG00000266979 LINC01180 3985 17 lncRNA 0.005521 1.18E-05 13 3.0 1 0.1 5.00E-06 4.00E-05
ENSG00000272070 AC005618.1 3147 5 lncRNA 0.006991 4.19E-05 18 4.2 1 0.1 2.39E-08 2.63E-07
ENSG00000251562 MALAT1 8828 11 lncRNA 0.005551 0.000283 30 7.0 19 2.2 4.19E-05 2.09E-04
ENSG00000261094 AC007066.2 2710 9 lncRNA 0.008118 0.000287 11 2.6 2 0.2 1.86E-04 7.43E-04
ENSG00000263874 LINC00672 4216 17 Protein_coding 0.005455 0.000493 14 3.3 2 0.2 9.70E-06 6.79E-05
ENSG00000180764 PIPSL 3349 10 Transcribed_processed_pseudogene 0.006569 0.0013 14 3.3 34 3.9 0.640 0.640
ENSG00000237984 PTENP1 3995 9 Transcribed_processed_pseudogene 0.005507 0.0013 15 3.5 22 2.5 0.376 0.640
ENSG00000240859 AC093627.4 5916 7 lncRNA 0.006085 0.0048 18 4.2 2 0.2 1.66E-07 1.66E-06
Fifteen noncoding genes with an enriched mutation rate compared to surrounding nonannotated regions. P-values are derived from the Fisher exact test (two-sided) and corrected for multiple testing using the FDR (Hochberg) method.
aENSG not recognised by ICGC data portal.
To further investigate the mechanisms underlying the observed SNVs and MNVs, we used the latest COSMIC mutational signatures (v3) to establish the presence and contribution of these predefined mutational signatures in metastatic CRC17. We identified 11 single base signatures (SBS) and 9 double base signatures (DBS) that had a relative contribution of at least 10% in minimally 10 cases and as such were considered dominant signatures in mCRC; SBS1, SBS5, SBS8, SBS9, SBS17b, SBS18, SBS35, SBS39, SBS40, SBS41, SBS44, DBS2-9, and DBS11. De novo signature calling using the Non-negative Matrix Factorization algorithm (NMF)18 did not identify additional signatures besides the known COSMIC signatures in our cohort.
Effects of systemic prior treatment on the genomic landscape
Patients receiving prior systemic treatment (n = 284) showed a significantly higher TMB, a higher number of SVs, a higher number of affected GISTIC CNV regions (7.58 versus 5.82; 208 versus 148; 31 versus 28, respectively; MWU p-values < 0.005), and more frequent occurrence of chromothripsis (6.5 versus 13.4%; Fisher exact test p = 0.042) compared to patients (n = 124) without prior systemic treatment. More specifically, we observed altered relative contributions for several mutational signatures in defined prior-treatment groups compared to treatment-naive patients (n = 124, Fig. 3 and Supplementary Data 2; MWU, FDR p < =5.15E-7). Patients who were prior-treated with a combination therapy of PLAT/PYR + target showed increased relative contributions of SBS8, SBS17b, SBS35, and DBS5 compared to treatment-naive patients. These results are supported by previous studies in which DBS5 and SBS35 signatures were linked to the effect of platinum (PLAT) compounds, while SBS17b was detected specifically in 5-FU or capecitabine (PYR) exposed tumors19. SBS8 was previously indirectly associated with prior platinum treatment in metastatic breast cancer17,20.Fig. 3 Mutational signatures in prior-treated cases compared to untreated cases.
Relative contribution (%) of several single and double base mutational signatures (SBS/DBS) in patients receiving prior treatment with platinum, pyrimidine antagonist, and targeted anti-EGFR treatment (PLAT/PYR + target; orange, n = 134) compared to untreated patients (blue, n = 124). Horizontal lines indicate the median. P-values are derived from the MWU test (two-sided) and corrected for multiple testing using the FDR (Hochberg) method. Source data are provided as a Source Data file.
Remarkably, even though TMB was increased in patients who received prior treatment compared to treatment-naive patients, no specific mutations (coding or noncoding) were associated with any of the defined prior-treatment groups or with prior treatment in general. With regard to the GISTIC-defined CNVs, we found increased frequencies of gains at 6p22.1, 6p21.1, and 18p11.32 as well as losses at 3p14.2 and 8p21.3 in patients who received prior treatment (Supplementary Table 2; chi-square FDR < 0.05). More specifically, gains of 6p22.1 and 6p21.1 were also associated with a prior-treatment regimen containing PLAT/PYR ± target whereas loss at 8p21.3 was only associated with PLAT/PYR + target.
Comparing metastatic CRC to primary CRC
The above described characteristics of our metastatic cohort were related to previous reports on primary CRC to identify changes potentially linked to the metastatic process (Supplementary Table 1). Therefore, we compared the observed relative contributions of the 20 dominant mutational signatures in our cohort to primary CRC data described by Alexandrov et al. (PCAWG cohort)17. For this analysis only the 124 untreated metastatic CRC cases from our cohort were included, since multiple treatments are known to specifically affect these mutational signatures17,19,20. SBS1, 8 and 41, as well as DBS2, 4, and 6 showed a significantly increased relative contribution in untreated metastatic cases (MWU, FDR ≤ 0.01; Fig. 4), suggesting they may be associated with the metastatic process. Etiologies for these signatures are either unknown (SBS8/41, DBS1) or appear age-related (SBS1, DBS2/DBS4), although DBS2 has also been linked to exposure to tobacco smoking and other endogenous and exogenous mutagens. Mutation frequencies per gene were compared between primary CRC (TCGA-DFCI cohort) and our total metastatic cohort. For this purpose, we selected genes mutated in primary CRC (TCGA-DFCI cohort) with >5% prevalence and complemented these with here identified metastatic driver genes regardless of their prevalence in primary CRC. Increased frequencies were only observed in driver genes TP53, ZFP36L2, KRAS, and APC (Fisher exact test, FDR ≤ 0.012). A decreased frequency was observed for 21 non-driver genes (Supplementary Table 3) and 1 driver gene, namely PIK3CA (Table 2). With respect to the identified putative noncoding drivers (Table 3), all of them were enriched in mCRC compared to primary CRC, except for PIPSL and PTENP1 (ICGC dataset; Fisher exact test, FDR < 5.74E-4).Fig. 4 Mutational signatures in primary CRC and untreated metastatic CRC.
Relative contribution (%) of several single and double base mutational signatures (SBS/DBS) in primary CRC tumors (purple, n = 73)17, compared to untreated metastatic CRC tumors (green, n = 124). Horizontal lines indicate the median. P-values are derived from the MWU test (two-sided) and corrected for multiple testing using the FDR (Hochberg) method. Source data are provided as a Source Data file.
Distinct mutational signature patterns in mCRC patients
Unsupervised hierarchical clustering using the 20 dominant mutational signatures complemented with mutational signatures previously described in primary CRC (SBS15/17a/28/37 and DBS10), and mutational signatures showing a dominant relative contribution (>25%) in at least one of our samples (SBS10a/10b), revealed three major and three minor groups of patients (Fig. 5).Fig. 5 Unsupervised hierarchical clustering of metastatic CRC using relative contribution of preselected mutational signatures.
Heatmap representing the median-centered relative contribution of mutational signatures between samples. Values were scaled from red (relative contribution above median) to yellow (relative contribution below median). Included single and doublet base signatures (SBS/DBS) are indicated at the right to which etiologies are added when known. Grouping of samples is shown by the dendrogram at the top. Source data are provided as a Source Data file.
The three major groups are found in cluster 1, cluster 3, and cluster 6. Clusters 1 and 6 are labeled “prior treatment” and “primary-like” as they are enriched for either patients with or without prior treatment compared to all other clusters (Fisher’s exact test: p = 4.588E-25 and p = 4.754E-15, respectively) and are characterized by higher relative contributions of signatures related to prior treatment (SBS5/8/35/17a/17b and DBS5) and signatures known from primary CRC (SBS1/5/18/40, DBS9), respectively. Samples from Cluster 6 are enriched (Fisher’s exact p = 0.005) for samples with >5% contribution of the recently described E. coli mutational signature in CRC as well21. Cluster 3 was labeled ‘mCRC-specific’ as it contains both patients with (n = 63) and without (n = 31) prior treatment characterized by higher relative contributions of signatures SBS9/37/39/41, which, except for SBS37, are rarely detected in primary CRC. Etiologies for SBS37/39/41 are unknown, whereas SBS9 mutations have been partly associated with polymerase eta (Pol η) function during somatic hypermutation in lymphoid cells. In vitro, Pol η activity has been associated with anticancer drugs resistance, specifically cisplatin and 5-FU22–24. Indeed we find that the majority of patients (13 out of 15) in cluster 3 with a high SBS9 contribution (≥10%) had already received prior treatment, although this did not reach statistical significance (Fisher’s exact test p = 0.07).
The remaining minor groups are found in Clusters 2, 4, and 5. Samples in clusters 2 and 4 are defined by a large contribution of DBS8 and DBS2, respectively. Cluster 5, labeled ‘high TMB’, contains 14 samples, which were all characterized by a high TMB (defined as >10/Mb) compared to only 82 out of the 415 remaining samples (20%) in the other clusters. High contributions of DNA mismatch repair associated signatures SBS15/44 and DBS7 characterize the 13 MSI samples in this cluster, whereas the one remaining sample showed high contributions of SBS10a/b, associated with polymerase epsilon (POLE) mutations.
MSI-specific gene mutations
We subsequently investigated whether specific somatic gene mutations were associated with each of the six clusters described above and found this was true only for the high TMB cluster (cluster 5). To correct for the higher likelihood of finding any mutation in a high TMB sample, we applied a permutation test25,26, which identified 28 genes as significantly more frequently mutated in the high TMB cluster versus all other samples (Fisher exact test, FDR and permutation p < 0.05, see Supplementary Table 4). As these 28 genes are large (cDNA size range 1.5–22 kb) and often contain substantial numbers of microsatellites and mononucleotide stretches (range 4–126), we evaluated whether their observed mutation frequency in MSI cases was significantly higher than the frequency distribution observed for all other genes with a comparable number (±10%) of MSI-prone coding sequences. Except for TNXB, for which we were unable to establish a reliable control distribution, all identified genes were significantly more frequently mutated in MSI cases compared to control genes containing similar numbers of MSI-prone sequences (one sample sign test; all p ≤ 0.0001). These results suggest that mutations in these genes are selected for during the disease process in MSI tumors. The top 2 genes, ACVR2A and UBR5, are known targets of the MSI process27. LRP1 mutations were found to reduce its expression in CRC and were associated with MSI status and poor outcome28. Although the other 25 identified genes were not previously associated with MSI status, three of these genes (KMT2C, KMT2D, and FAT1) were present in the Yaeger dataset of mCRC samples6. Mutations in all three overlapping genes were significantly enriched in MSI cases (n = 16) compared to microsatellite stable (MSS) cases (n = 305) in this dataset as well (all Fisher p < =9.19E-7).
Association between molecular landscape and treatment response
The observed molecular characteristics were associated with response to current treatment for the 286 patients in our cohort with recorded treatment response. These results should be interpreted with caution due to the heterogeneity of our cohort in terms of both treatment line and type of prior treatments received, which may introduce bias. We studied ordinal response (PD, SD, and PR) to any treatment as well as to specific treatment regimens. In total, 123 items were used as input in the regression model, consisting of five themes (full list in Supplementary Data 3): clinical parameters (age, gender, prior treatment, and radiotherapy), counts (TMB, kataegis, chromothripsis, total number of SV by type and the number of 10kb–1Mb deletions), mutational signatures (DBS/SBS), driver genes (including noncoding genes), and GISTIC-defined CNVs. Items that reached univariate statistical significance (p < 0.05) were used in a multivariable penalized ordinal regression model for treatment response (Table 4).Table 4 Multivariate LASSO analysis.
Type Item All treatments Oxaliplatin containing PLAT/PYR Target-mono TOP/PYR
Clinical Prior Treatment 0.57
Gender 0.77 −1.14 −1.10
Counts nr of Tandem Duplications 1.19
nr of 10kb–1Mb deletions 0.01
Mutational Signatures DBS2 −0.02
DBS5 0.13
DBS11 −0.03
SBS17b 0.07 0.13
SBS39 0.04
SBS41 −0.21
Driver Genes APC 0.23
KRAS 0.78
PIK3CA 0.22
FBXW7 17.65
Non-coding LINC00672 0.90
GISTIC Regions Gain 17q12 (ERBB2*) 0.44
Gain 18p11.32 (CETN1*) 0.59
Gain 20q11.1 (BCL2L1*) 0.12
Gain 8p11.21 (KAT6A) −0.78 −1.56
Gain 7p21.3 (VWDE) 3.68
Gain 7q31.2 (MET*) 3.59
Gain 7p12.3 (PKD1L1) 3.04
Gain 7q34 (no genes in peak) 3.59
Gain 14q23.1 (no genes in peak) 1.64
Loss 18q12.2 (hsa-mir-924*) −1.52 −3.30 2.79
Loss 6q26 (PARK2) −1.38
Loss 9p21.3 (CDKN2A*) −1.88
Loss 16q23.1 (WWOX) −1.70
Loss 4q22.1 (CCSER1) 2.66
Loss 4q35.1 (IRF2) 1.78
Loss 18q21.2 (SMAD4) −1.35 −3.30 2.79
Loss 18q23 (NFATC1*) −1.14 −3.30
Loss Xp22.31 (STS*) −1.40 1.78
Loss 14q23.3 (GPHN) −2.09
Items that reached univariate statistical significance (p < 0.05) were used in a multivariable penalized ordinal regression model for treatment response. Univariate regression was performed for genomic features (Supplementary Data 3) using the ‘polr’ function from the MASS R package (v7.3-51.4) and subsequently those with a univariate two-sided p-value <0.05 were selected for multivariable ordered LASSO regression using the ordinalNet R package (v2.7). Regression coefficients are shown for features that remained significant in the multivariable model.
*Multiple genes present in region; bold: known Fragile Site region
Overall we found that, next to receiving prior treatment(s), the number of 10kb–1Mb deletions, mutations in KRAS, APC, PIK3CA, and LINC00672, mutational signatures SBS17b/39, DBS2/5/11, and gains at 18p, 17q, and 20q were associated with treatment response regardless of treatment type in mCRC patients. For SBS17b this effect was more pronounced when specifically investigating patients treated with platinum as described before17. CNVs were predominantly associated with response to PLAT/PYR or PYRmono treatment, whereas mutations in FBXW7 were associated with poor response to targeted treatment. FBXW7 mutations were detected in 51 patients from our cohort, including 21 KRAS wild-type patients. Of these 21 patients, five were treated with panitumumab monotherapy, all of whom had PD as best response. This suggests that, next to somatic KRAS mutations, somatic FBXW7 mutations may provide an additional negative selection marker for anti-EGFR treatment. This finding is in concordance with previous reports on FBXW7 mutation prevalence in nonresponding patients on anti-EGFR treatment29,30.
Potential clinical implications
WGS data of our cohort of 429 patients with metastatic CRC revealed several potential molecular features that might be associated with sensitivity to particular anticancer agents. A high TMB (here defined as >10 mutations per Mb) has been suggested as a potential selection tool for tumors that may respond to immunotherapy31. In our cohort, 96 (22%) samples showed a TMB > 10, of which 13 were MSI. A gradual increase in TMB was observed with the number of prior treatments (test for trend, p = 4.39E-13). For the subset of samples of which we also had RNA-seq data available, we calculated the Tumor Infiltrating Leukocyte (TIL) score as a proxy for the immunogenicity of the tumor32. Interestingly, we did not observe a significantly higher TIL score in the TMB-high samples (n = 21) compared to the other samples (n = 63; MWU; p = 0.39), whereas the average TIL score in MSI samples is significantly higher compared to both MSS samples with a high TMB and with a low TMB (Kruskal–Wallis test (p = 0.037) followed by Dunn’s pairwise comparison (Benjamini–Hochberg corrected p = 0.012 and p = 0.021 for MSI compared to MSS with high and low TMB, respectively (See Supplementary Fig. 2). Although far from definite, these results support the on-label use of immunotherapy in MSI tumors and suggest that merely using TMB may not be sufficient to identify the tumors with immunogenic potential in the metastatic setting.
Other on-label markers found in our cohort include a targetable BRAF V600E mutation in 40 patients, as well as 130 RAS/RAF wild-type patients that did not receive targeted anti-EGFR treatment yet. However, our data suggest that mutations in FBXW7, observed in 21 out of these 130 RAS/RAF wild-type patients, should be considered as a contra-indication for the use of anti-EGFR treatment. Molecular biomarkers for potential off-label use that were found in our cohort include amplifications of ERBB2 (HER2), MET and CDK4, loss of BRCA1 and BRCA2 through deletion or high impact mutations, loss of TSC1 and TSC2 through high impact mutations, and possible fusions of PDGFRB. In addition, 23 patients in our cohort carried a KRAS G12C mutation, for which an inhibitor may become available in the near future33.
In summary, for 55% of our patients one or more targeted treatments are potentially available based on the molecular profile of their cancer (Fig. 6).Fig. 6 Actionable genes.
Data from OncoKB were matched to affected genes observed in our mCRC cohort. Numbers indicate the number (and percentage) of affected patients. Source data are provided as a Source Data file.
Discussion
This study encompasses a WGS-based, comprehensive description of the molecular landscape of metastatic CRC and aims to put this landscape into perspective by associating it with prior systemic treatments, comparing it to primary CRC and relating it to treatment response.
In general, the genomic landscape of CRC remains relatively stable in metastatic disease. However, compared to primary CRC, our metastatic CRC cohort showed significant enrichment for mutations in 4 out of 23 coding and 12 out of 15 noncoding (putative) driver genes. From the identified putative drivers, only mutations in PIK3CA were significantly decreased in mCRC. Six of our identified coding driver genes are not present in the current CRC-specific MSK-IMPACT panel, namely ZFP36L2, BCL, BCL9L, ELF3, LMTK3, and TGIF1.
Within the mCRC cohort we observed clear effects of received prior treatments on the total numbers of aberrations, CNVs, and mutational signatures, with the latter sufficiently dominant to show up as a separate group after hierarchical clustering. Remarkably, we also observed an mCRC-specific cluster characterized by signatures which are rarely found in primary CRC and are not associated with any treatment (SBS9/39/41). SBS9 is associated with Pol η activity, an error-prone polymerase encoded by the POLH gene, which mediates translesion synthesis and is induced by replication stress34. Interestingly, high levels of Pol η have been associated with cancer therapy resistance in vitro22–24. We did observe that the majority of patients with a high relative SBS9 contribution had already received prior treatment; however, unfortunately, sample numbers were too low to directly associate SBS9 contribution with POLH expression in our dataset. Another predominant cluster group consisted of metastatic MSI samples. In these samples we observed a significant enrichment of mutations in a specific set of genes compared to other similarly MSI-prone genes, suggesting these genes are preferentially affected or selected for during disease progression.
The varying number and types of received prior treatments within our cohort hampered the search for prognostic and predictive biomarkers. However, we found that, next to already known events, the number of LINC00672 mutations and 10kb–1Mb deletions were associated with treatment response irrespective of the type of treatment. Strikingly, many of these recurrent deletions occur in known Common Fragile Site (CFS) genes, as described in primary CRC as well13, implicating replication stress as one of driving mechanisms35. In addition, FBXW7 mutations were predictive for poor response to EGFR-targeted treatments in our prospective cohort. This is in line with previous observations showing that FBXW7 mutations were enriched in unresponsive patients compared to patients responding well to EGFR-targeted treatments29,30.
The current study gives a detailed description of the genomic landscape of metastatic CRC. More specifically, our study identifies treatment-induced changes, metastasis-specific alterations, and associations between molecular traits and response to treatment. In addition, we provide prospective validation for FBXW7 mutations as a predictive biomarker for poor response to EGFR-targeted treatment. Combined with future studies, this catalogue of molecular alterations will speed up the identification of resistance mechanisms, the determination of metastasis-driving processes, and, ultimately, the improvement of metastatic CRC patient care.
Methods
Patient cohort and study procedures
Colorectal cancer patients included in this study were selected from the previously described cohort of the Center for Personalized Cancer Treatment (CPCT) consortium (CPCT-02 Biopsy Protocol, ClinicalTrial.gov no. NCT01855477), which was approved by the medical ethics committee of the University Medical Center Utrecht, the Netherlands9. All patients have given explicit consent for whole-genome sequencing and data sharing for cancer research purposes. Upon our data request for all CRC patients thus far, we were provided with the data of all patients registered as metastatic CRC patients included between April 2016 and January 2019 (n = 487). Patients who received systemic treatment which is not normally given to colorectal cancer patients (e.g., carboplatin, paclitaxel, sunitinib, and etoposide) were excluded to avoid erroneous inclusion of patients suffering from another type of cancer (n = 28). When multiple biopsies were included for one patient (n = 29), only the first biopsy was included in our analyses. In total, we included 429 distinct CRC patients in our analyses. Based on the provided information regarding all forms of systemic treatment patients received before the study biopsy took place (further referred to as “prior treatment”), we coded the (groups of) active agents using the following abbreviations: PLAT (oxaliplatin), PYR (fluoropyrimidines), TOP (topoisomerase inhibitor; Irinotecan), +targeted (when bevacizumab or panitumumab was added), CHEMCOM (triplet combination therapy). Prior-treatment regimens were grouped based on their working mechanism to enable the analysis of their effect on the genomic landscape. Treatment related analyses were performed using combinations of the abbreviations mentioned above. For detailed information see Table 1.
Whole-genome sequencing; identification of somatic changes
Whole-genome sequencing of paired tumor/normal was performed in all cases. In short, raw sequencing data were processed using bcl2fastq (versions 2.17 to 2.20), mapped to the human reference genome GRCh37 using BWA-mem v0.7.Sa and GATK BQSR and Haplotype Caller v3.4.46 and Strelka v1.0.14 were used to call somatic mutations. Within our cohort, 98% of the biopsies of metastatic lesions showed a coverage of at least 30× (95% with >60× coverage), whereas for the normal blood 98% had >10× coverage and 94% >20× coverage. The identification of copy number changes was performed using GISTIC v2.0.2336 with the following parameters: genegistic 1; gcm extreme; maxseg 4000; broad 1; brlen 0.98; conf 0.95; rx 0; cap 3; saveseg 0; armpeel 1; smallmem 0; res 0.01; ta 0.1; td 0.1; savedata 0; savegene 1; and qvt 0.19,20.
RNA sequencing and CMS calling
Matched RNA was isolated from the same frozen tissue for 91 CRC patients on an automated setup (QiaSymphony) according to supplier’s protocols (Qiagen) using the QIAsymphony RNA Kit for tissue and quantified by Qubit. A total of 50–100 ng of RNA was used as input for library preparation using the KAPA RNA HyperPrep Kit with RiboErase (Human/Mouse/Rat) (Roche). Barcoded libraries were equimolarly pooled and sequenced using standard settings (Illumina) on either a NextSeq 500 (V2.5 reagents) generating 2 × 75 read pairs or a NovaSeq 6000 generating 2 × 150 read pairs. BCL output was converted to FASTQ using bcl2fastq (versions 2.17–2.20) using default parameters and sequence reads were trimmed for adapter sequences using fastp (v0.20.0). The resulting FASTQ files were mapped to GRCh38 using STAR (v2.6.1d)37. Sambamba (v0.7.0)38 was used to mark duplicates and index the resulting BAM files. Gene annotation was derived from GENCODE Release 30 (https://www.gencodegenes.org/), raw read counts were obtained with featureCounts (v1.6.3)39 and normalized using GeTMM40. Normalized data were used to (1) determine CMS with both the single-sample prediction parameter from the “CMSclassifier” package (v1.0.0) (https://github.com/Sage-Bionetworks/CMSclassifier)2 and CMSCaller v(0.99.1)11, and (2) calculate the Tumor Infiltrating Lymphocytes (TIL) score by averaging the expression of TIL-genes32
Identification of mutational signatures and driver genes
Mutational signatures (COSMIC v3)17 were called using R package MutationalPatterns v1.10.018, focusing on single and double base signatures. This package was also used to perform de novo signature calling using the Non-negative Matrix Factorization (NMF) method. Detection of kataegis and chromothripsis was performed as previously described41. In short, to call kataegis only SNVs were considered to establish segments based on the intermutational distance. Segments were determined using a piecewise constant fitting model and were called as kataegis when at least five SNVs were present showing an intermutational distance of ≤2 kb. Chromothripsis-like events were called using the Shatterseek R package (v0.4). Driver genes, i.e., genes under selective pressure, were identified by the dN/dS model using R package dndscv (v0.0.1.0)42. A global q ≤ 0.05 was used to select statistically significant driver genes. The R package discover v0.9.243 was used to test for mutual exclusivity. To identify noncoding genes with an enriched mutation rate, we first established a baseline mutation rate based on all identified SNVs, MNVs and Indels found in nonannotated regions, as we assume these regions are not under any selective pressure. Nonannotated regions were based on GENCODE annotation (version33) and for each of these regions we calculated a mutation rate (number of mutations/size of region). Next, a mutation rate (number of mutations/size of noncoding gene) was calculated for all somatic mutations annotated as ‘noncoding transcript variant’. The signed-rank test with Hochberg’s multiple testing correction was used to determine whether the mutation rate of a recurrent noncoding gene (mutated in at least 5% of the cohort) exceeded the baseline mutation rate. Per noncoding gene a specific baseline was determined using only nonannotated regions (>1 kb) in an area of 2 Mb surrounding the respective noncoding gene.
Verification in publicly available datasets
To compare mutational signatures, publicly available WGS data from 73 primary colorectal cases were used17. We downloaded the matrix of counts for single and double base substitutions of primary cases and analyzed these in the same manner as the metastatic CRC cases to call mutational signatures. Observed frequencies of mutated genes in metastatic CRC were verified and compared in two publicly available datasets. Dataset 1, the Yaeger dataset, contained 321 unique metastatic CRC patients that were profiled for mutations by targeted sequencing6. Dataset 2, the TCGA-DFCI dataset, contained 1949 unique primary CRC patients that were profiled for mutations in coding regions (accessed via cBioPortal January 21, 2020). Prior to analysis, synonymous mutations were removed and multiple mutations within the same gene were aggregated per patient. Dataset 3, the ICGC dataset, was used to compare mutation frequencies of noncoding genes and contained 866 unique primary CRC patients with available mutation data (accessed via the ICGC data portal, release 28). The used cohorts are summarized in Supplementary Table 1.
Estimating MSI-prone sequences
To evaluate preferentially mutated genes in MSI cases, the number of MSI-prone sequences in a gene are of interest. Data of the Microsatellite Database (MSDB, https://data.ccmb.res.in/msdb/, June 2, 2020) were filtered for repeats annotated to human exons44. For each gene, the number of repeats was summed. In addition, a custom Perl script was used to count mononucleotide stretches of lengths between 6 and 13 (the latter is the minimum length used in MSDB) as we noticed many InDels in our data in mononucleotide stretches less than 13 bases long. Exon sequences of the Consensus CDS database (https://www.ncbi.nlm.nih.gov/CCDS/) were used to count the number of mononucleotide stretches per gene.
Associations with response to treatment
Treatment response was evaluated according to RECIST (v.1.1) every 8 to 12 weeks depending on the treatment regimen and was defined as response (partial or complete), stable disease, or progressive disease45. For regression analyses, the best overall response was used as outcome measure. Genomic features (at least 5 events per group) were associated with response to treatment in a 2-step procedure using ordinal LASSO (least absolute shrinkage and selection operator) regression, which is suited for datasets with a relatively high number of predictors in comparison to cases and protects against overfitting. First, univariate regression was performed for genomic features (Supplementary Data 3) using the ‘polr’ function from the MASS R package (v7.3–51.4) and subsequently those with a univariate p-value <0.05 were selected for multivariable ordered LASSO regression using the ordinalNet R package (v2.7).
Identification of potentially actionable events
OncoKB (accessed on March 31, 2020) was used to identify clinically actionable genes from the list of mutated genes in our cohort, using only genes with level 1 and 2 evidence46. In case OncoKB listed a specific gene alteration as actionable genomic aberration, we only counted patients that harbored that specific mutation or CNV. For genes for which only ‘Oncogenic mutations’ were listed by OncoKB, we only included patients if the gene had a mutation with ‘High impact’ consequence (i.e., a nonsense or frameshift mutation). To evaluate patients eligible for an anti-EGFR therapy, we included only patients that were triple wild-type for KRAS, NRAS, and BRAF, and excluded those patients that had already received anti-EGFR therapy prior to biopsy.
Statistics
In general, a Pearson’s Chi-squared test or Fisher’s exact test (in case of too few expected events) was used to evaluate the categorical data while continuous variables were evaluated using either a Mann–Whitney U test (MWU) or a Kruskal–Wallis H (KWH) test depending on the number of categories. All statistical tests were two-sided and considered statistically significant when P < 0.05. Stata 13.0 (StataCorp) and R (v3.6.0) were used for the statistical analyses. Multiple testing using the Hochberg procedure to correct P values was applied when necessary. The statistical test used is specified throughout the results section.
Reporting summary
Further information on research design is available in the Nature Research Reporting Summary linked to this article.
Supplementary information
Supplementary Information
Peer Review File
Description of Additional Supplementary Files
Supplementary Data 1
Supplementary Data 2
Supplementary Data 3
Reporting Summary
Source data
Source Data
Supplementary information
Supplementary information is available for this paper at 10.1038/s41467-020-20887-6.
Acknowledgements
We thank the Hartwig Medical Foundation, and Stichting Stelvio for Life for financial support of clinical studies and WGS analyses. We thank the Center for Personalized Cancer Treatment for proving the clinical data. We would like to thank all local principal investigators, medical specialists, and nurses of all contributing centers for their help with patient accrual. We are particularly grateful to all participating patients and their families.
Author contributions
P.A.J.M., M.S., S.S., J.W.M.M., and S.M.W. wrote the manuscript, which all authors reviewed. M.S. and J.V.R. performed the bioinformatics analyses. P.A.J.M., L.A., and S.S. managed clinical data assessment. M.L., N.S., M.P.H., G.A.C., J.M.V.R., and A.J.T.T. are main clinical contributors. M.P.L. and S.S. are members of the CPCT-02 study team and/or CPCT board. E.C. coordinated the sequencing of samples and contributed to the bioinformatics analyses.
Data availability
The WGS, RNA-seq, and corresponding clinical data used in this study was made available by the Hartwig Medical Foundation (Dutch nonprofit biobank organization) after signing a license agreement stating data cannot be made publicly available via third party organizations. Therefore, the data are available under restricted access and can be requested upon by contacting the Hartwig Medical Foundation (https://www.hartwigmedicalfoundation.nl/applying-for-data/) under the accession code DR-058. Publicly available datasets that were used in this study are listed in Supplementary Table 1. The Yaeger data used in this study are available in the cBioPortal for Cancer Genomics (http://www.cbioportal.org/study?id=crc_msk_2017). The TCGA-DFCI data used in this study have been deposited in the cBioPortal for Cancer Genomics which we accessed on January 21, 2020 (https://www.cbioportal.org/study/summary?id=coadread_tcga; https://www.cbioportal.org/study/summary?id=coadread_tcga_pub; https://www.cbioportal.org/study/summary?id=coadread_tcga_pan_can_atlas_2018; https://www.cbioportal.org/study/summary?id=coadread_dfci_2016). The ICGC data used in this study have been deposited in the ICGC data portal (release 28) (https://dcc.icgc.org/projects/COAD-US; https://dcc.icgc.org/projects/COCA-CN; https://dcc.icgc.org/projects/READ-US). The remaining data are available within the Article, Supplementary Information or available from the authors upon request. Source data are provided with this paper.
Code availability
The bioinformatical code used for data processing is available at https://github.com/hartwigmedical/pipeline5.
Competing interests
The authors declare no competing interests.
Peer review information Nature Communications thanks Kin Chan, Silvia Marsoni and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
These authors contributed equally: Pauline A.J. Mendelaar, Marcel Smid.
Change history
5/27/2021
A Correction to this paper has been published: 10.1038/s41467-021-23629-4 | PANITUMUMAB | DrugsGivenReaction | CC BY | 33495476 | 18,948,086 | 2021-01-25 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Therapy partial responder'. | Whole genome sequencing of metastatic colorectal cancer reveals prior treatment effects and specific metastasis features.
In contrast to primary colorectal cancer (CRC) little is known about the genomic landscape of metastasized CRC. Here we present whole genome sequencing data of metastases of 429 CRC patients participating in the pan-cancer CPCT-02 study (NCT01855477). Unsupervised clustering using mutational signature patterns highlights three major patient groups characterized by signatures known from primary CRC, signatures associated with received prior treatments, and metastasis-specific signatures. Compared to primary CRC, we identify additional putative (non-coding) driver genes and increased frequencies in driver gene mutations. In addition, we identify specific genes preferentially affected by microsatellite instability. CRC-specific 1kb-10Mb deletions, enriched for common fragile sites, and LINC00672 mutations are associated with response to treatment in general, whereas FBXW7 mutations predict poor response specifically to EGFR-targeted treatment. In conclusion, the genomic landscape of mCRC shows defined changes compared to primary CRC, is affected by prior treatments and contains features with potential clinical relevance.
Introduction
Primary colorectal cancer (CRC) can be divided into a major group of chromosomally instable tumors and a minor group of hypermutated, chromosomally stable tumors due to microsatellite instability (MSI) or POLE mutations1. Parallel to the described genomic subtype division, transcriptomic analysis was used to identify four consensus molecular subtypes (CMSs) with distinguishing features including prognosis2.
Molecular analysis of CRC revealed specific genetic alterations with clinical implications. Mutations in KRAS and BRAF predict failure to treatment with EGFR-inhibitors, whereas copy number alterations of ERBB2 or IGF2, and the occurrence of chromosomal translocations leading to fusion genes such as NAV2/TCF7L1, are potentially drug targetable1,3.
Although the molecular knowledge of primary CRC has contributed to a better understanding of its pathogenesis, cancer-related mortality usually occurs as a consequence of distant metastases, in which ongoing mutational processes and selective treatment pressure can result in altered molecular characteristics4.
To date, in-depth analyses of large series of colorectal cancer metastases are limited to studies using either whole-exome sequencing (WES) or targeted sequencing of cancer-associated genes4–6. Although these studies yielded extensive knowledge on the presence of specific genomic aberrations in mCRC, they do not necessarily reflect its complete molecular landscape. For optimal identification of mutational signatures, the power provided by whole-genome sequencing (WGS) data greatly exceeds that of WES7. Next to this, WGS simultaneously allows for the determination of MSI, structural rearrangements, chromothripsis, and kataegis. In addition, clinically relevant genetic alterations within noncoding regions were recently reported in primary CRC8. To date, the only other study which reported in detail on WGS data of colorectal metastases included 12 patients4.
Here, we provide a comprehensive description of the molecular landscape of metastatic CRC (mCRC). We use WGS data obtained from a large multicenter, prospective collection of snap-frozen metastatic tissue biopsies from 429 patients starting a new line of systemic treatment9. In addition, matched RNA-seq data are available for 91 patients. The observed metastatic molecular landscape is compared to WGS data of primary CRC cohorts (Supplementary Table 1), associated with prior treatments as well as treatment response, and evaluated for clinical utility.
Results
Cohort description
Clinical characteristics of our included cohort of 429 patients are summarized in Table 1. Median tumor purity (0.53 (IQR 0.38–0.67) was estimated on the obtained sequencing data and was not significantly different between biopsy sites. Based on a previously described WGS data analysis algorithm9 14 samples (3%) were scored as microsatellite instable (MSI), which is in concordance with the observed MSI frequency in mCRC in literature (4%)10.Table 1 Cohort description.
Patient details Number of patients
Total cohort 429
Gender Female 174
Male 255
Age (median (IQR,Range)) 64 (IQR 56–72, range 25–88)
Prior-treatment details
Systemic prior treatment: yes 284
Treatment regimen Manuscript code
5-FU/capecitabine–oxaliplatin doublet (CAPOX, FOLFOX) PLAT/PYR 121
+bevacizumab PLAT/PYR + targeted 134
5-FU—Topisomerase inhibitor doublet (Irinotecan based, FOLFIRI) TOP/PYR 26
+bevacizumab/panitumumab TOP/PYR + targeted 9
5-FU/capecitabine monotherapy PYRmon 39
+bevacizumab PYR + targeted 36
Topoisomerase inhibitor (Irinotecan) monotherapy TOPmono 67
+bevacizumab TOP + targeted 7
Oxaliplatin + bevacizumab/panitumumab PLAT + targeted 5
Panitumumab/cetuximab/encorafenib+binimetinib/bevacizumab/regorafenib Targeted mono 35
5-FU/capecitabine–oxaliplatin–irinotecan triplet (FOLFOXIRI) CHEMCOM 2
+bevacizumab CHEMCOM + targeted 5
Other Diverse 15
Systemic prior treatment: no 124
Systemic prior treatment: unknown 21
Radiotherapeutic prior treatment: yes 109
RT + systemic treatment 68
Chemoradiation 33
Radiotherapy only 8
Radiotherapeutic prior treatment: no 299
Radiotherapeutic prior treatment: unknown 21
Biopsy details Number of patients
Origin
Liver 287
Soft tissue 84
Lymph node 24
Lung 21
Other 13
Technical details
Tumor purity 0.53 (IQR 0.38–0.67)
Read coverage (median) 102.6× (IQR 94.6×–112.0×)
Patient Characteristics. PLAT/PYR (5-FU/capecitabine–oxaliplatin doublet (CAPOX, FOLFOX)), PLAT/PYR + targeted (bevacizumab added), TOP/PYR + targeted (bevacizumab added), PYRmono (5-FU/capecitabine monotherapy), PYR + targeted (bevacizumab added), TOP/PYR (5-FU-Topisomerase inhibitor doublet (Irinotecan based)), TOPmono (Topoisomerase inhibitor (Irinotecan) monotherapy), TOP + targeted (bevacizumab added), PLAT + targeted (Oxaliplatin + bevacizumab/panitumumab), Targeted mono (panitumumab, cetuximab, encorafenib + binimetinib, bevacizumab, regorafenib), CHEMCOM (5-FU/capecitabine–oxaliplatin–irinotecan triplet (FOLFOXIRI)), CHEMCOM + targeted (bevacizumab added), Other (diverse).
Based on the treatment data, the cohort can be divided in patients who did (n = 284) and who did not (n = 124) receive any systemic treatment prior to the moment the biopsy was taken. Within the group of prior-treated patients, 13 different combinations of treatment regimens were defined as specified in the materials and methods and listed in Table 1.
For 91 cases RNA-seq data were available, allowing us to determine their Consensus Molecular Subtype (CMS). Remarkably, using the CMS-classifier package, none of the metastatic CRC samples were classified as CMS3, whereas 10 were classified as CMS1, 41 as CMS2, and 14 as CMS4. The remaining 26 samples (29%) could not be classified into one of the 4 subtypes, which might be partly due to the presence of normal cells of noncolon origin in our metastatic setting. Indeed, using the alternative CMSCaller algorithm, which is less dependent on signals from the tumor microenvironment, reduced the number of unclassified samples to 14 (15%), whereas still only 3 samples were classified as CMS311. Twenty-two samples were classified as CMS1, 25 as CMS2, 3 as CMS3, and 27 as CMS4.
Regardless of the calling algorithm used, the estimated tumor cell percentage was significantly lower in biopsies classified as CMS4 than in the other subtypes (medians CMS1: 52.5 and 45%; CMS2 61 and 61%; CMS3: none and 66% and CMS4: 34.5 and 42%; KWH; p = 0.0007 and p = 0.0156 for CMS Classifier and CMSCaller, respectively), which is concordant with the described high-stroma content in this subtype2.
The molecular landscape of mCRC
From the WGS data of all 429 cases, we distilled somatically acquired single nucleotide variants (SNVs), multiple nucleotide variants (MNVs), structural variants (SVs), insertions/deletions (InDels), and copy number variants (CNVs). The overall tumor mutational burden (TMB) representing the amount of SNVs, MNVs and InDels per Megabase (Mb), ranged from 0.96 to 366.15 with a median of 7.01 (95% CI 6.62–7.47). Using GISTIC2.0, we identified 55 recurrent CNVs (29 gains and 26 losses) within our entire cohort, containing a number of already known and putative driver genes (Supplementary Data 1). Chromothripsis was observed in 47 cases (11%), whereas kataegis was observed in 102 cases (24%), involving just a single chromosomal region in two-third of cases, with a maximum of 10 regions in one single case. Presence of kataegis was associated with MSI and high TMB (≥10; test for trend p = 0.00014). In fact, 9 out of 13 MSI cases had at least two kataegis regions.
We further evaluated the type and size of SVs observed in our cohort (Fig. 1). A broad range of differently sized Tandem Duplications (TD; ~14–93 kb) with a peak at 26 kb was observed, which was clearly distinct from the TD sizes previously observed in other cancers (~11 kb in BRCA1-mutated, ~231 kb in CCNE1-activated, and ~1.7 Mb TDs in CDK12-mutated cancer, respectively)12. Inversions in mCRC are usually over 10 Mb in size, while deletions range from ~10 kb to 1 Mb, with a distinct peak at ~128 kb. Events within this latter peak include many recurrent deletions in known Common Fragile Site (CFS) genes: e.g., FHIT, RBFOX1, and MACROD2. This phenomenon involving frequent deletions of CSF genes was recently described in primary CRC as well13.Fig. 1 Size distributions of the different types of structural variants.
Ridge-plot of the density of genomic sizes of structural variants in metastatic CRC. INV inversions (blue), DUP tandem duplications (purple), DEL deletions (orange). Source data are provided as a Source Data file.
Using the ratio of nonsynonymous to synonymous substitutions caused by the somatic nucleotide mutations (SNV and InDels; dN/dS analysis), 23 genes were identified as putative driver genes (q < 0.05, Fig. 2, Table 2). In 99.1% of cases (425 out of 429) at least one of these 23 putative driver genes was mutated. Testing for mutual exclusivity only revealed already known associations: KRAS with BRAF/NRAS/RNF43/TP53 (q = 1.06E-7, q = 1.54E-4, q = 0.004, and q = 0.017, respectively), and APC with RNF43/BRAF (both q = 1.54E-4; Supplementary Fig. 1). For those genes also present in the targeted panel used by Yaeger et al.6, comparable mutation frequencies were observed in both cohorts (Table 2).Fig. 2 Oncoplot of metastatic CRC depicting identified driver genes and somatic mutations (SNV, InDels, and MNV).
Top panel: genes identified by dN/dS as driver genes per type of mutation; purple: frameshift variant; orange: other variant; blue: stop/gain variant; green: structural variant. Bottom panel: first track: clinical information: sex (male: orange; female: green) and second track: biopsy site. Track three (PLAT/PYR ± targeted) indicates which patients have been treated with platinum-based therapy (PLAT; e.g., oxaliplatin) and a pyrimidine-targeting drug (PYR; e.g., 5-FU), with or without the addition of another targeted treatment (±targeted; e.g., bevacizumab). Tracks four to six depict the distribution of the consensus molecular subtypes (CMS), tumor mutational burden (TMB), and the number of structural variant deletions of size 10kb–1Mb (DEL_CFS), partly associated with Common Fragile Sites (CFS), respectively. Source data are provided as a Source Data file.
Table 2 Mutation frequency driver genes.
Metastatic CRC Primary CRC—(TCGA, combined studies cBioportal) Metastatic CRC—(Yaeger et al.)
Gene dN/dS q-value Mutations (N) Mutations (%) Mutations (N) Mutations (%) Fisher p-value FDR Hochberg % change in meta Mutations (N) Mutations (%) Fisher p-value FDR Hochberg
TP53 0 317 73.9 1123 57.6 2.04E-10 4.7E-09 16.3 246 76.6 0.395 1
ZFP36L2 0 42 9.8 97 5.0 3.61E-04 0.008 4.8 Not present
KRAS 0 203 47.3 744 38.2 5.88E-04 0.012 9.1 127 39.6 0.037 0.487
APC 0 336 78.3 1372 70.4 8.86E-04 0.018 7.9 241 75.1 0.335 1
PIK3CA 0 68 15.9 445 22.8 0.001 0.023 −7.0 49 15.3 0.840 1
B2M 5.37E-03 8 1.9 91 4.7 0.007 0.128 −2.8 2 0.6 0.202 1
SMAD4 0 74 17.2 243 12.5 0.010 0.164 4.8 47 14.6 0.367 1
ATM 9.50E-04 33 7.7 227 11.6 0.017 0.266 −4.0 18 5.6 0.306 1
FBXW7 0 51 11.9 301 15.4 0.061 0.912 −3.6 25 7.8 0.068 0.814
AMER1 0 37 8.6 209 10.7 0.220 0.913 −2.1 11 3.4 0.004 0.067
ARID1A 1.13E-09 39 9.1 201 10.3 0.480 0.913 −1.2 15 4.7 0.022 0.333
BCL9 5.27E-02 28 6.5 107 5.5 0.419 0.913 1.0 Not present
BCL9L 2.24E-06 27 6.3 133 6.8 0.750 0.913 −0.5 Not present
BRAF 0 56 13.1 273 14.0 0.644 0.913 −1.0 38 11.8 0.657 1
ELF3 5.37E-03 7 1.6 51 2.6 0.299 0.913 −1.0 Not present
LMTK3 1.33E-02 15 3.5 56 2.9 0.530 0.913 0.6 not present
NRAS 0 26 6.1 125 6.4 0.913 0.913 −0.4 14 4.4 0.329 1
PTEN 4.30E-08 17 4.0 123 6.3 0.069 0.913 −2.3 14 4.4 0.854 1
RNF43 2.20E-02 28 6.5 162 8.3 0.239 0.913 −1.8 21 6.5 1.000 1
SMAD3 5.68E-03 11 2.6 72 3.7 0.309 0.913 −1.1 11 3.4 0.518 1
SOX9 0 41 9.6 177 9.1 0.782 0.913 0.5 16 5.0 0.025 0.352
TCF7L2 1.07E-09 50 11.7 177 9.1 0.103 0.913 2.6 19 5.9 0.007 0.116
TGIF1 1.33E-02 18 4.2 62 3.2 0.300 0.913 1.0 Not present
Twenty-three genes identified as putative driver genes using the ratio of nonsynonymous to synonymous substitutions caused by the somatic nucleotide mutations (SNV and InDels; dN/dS analysis). P-values are derived from the Fisher exact test (two-sided) and corrected for multiple testing using the FDR (Hochberg) method.
Similarly, for 15 noncoding genes an enriched mutation rate was observed compared to surrounding nonannotated regions (Table 3), suggesting these genes are relevant for the oncogenic process. These noncoding genes include PTENP1, a known tumor suppressor in CRC14, MALAT1, for which an increased mutation rate was already described in a pan-cancer analysis15, and LINC00672, described to promote chemo-sensitivity16.Table 3 Mutation frequency noncoding genes.
Metastatic CRC Primary CRC - (ICGC)
ENSG Symbol Size Chr Type Mutation rate FDR Hochberg Mutations (N) Mutations (%) Mutations (N) Mutations (%) Fisher p-value FDR Hochberg
ENSG00000273001 AL731533.2 577 10 lncRNA 0.067591 0 6 1.4 0 0 0.001291 3.87E-03
ENSG00000280325 AC074183.2 921 7 TEC 0.033659 0 25 5.8 a a
ENSG00000261584 AL513548.1 1723 6 lncRNA 0.012768 9.93E-24 14 3.3 3 0.3 3.84E-05 2.09E-04
ENSG00000259834 AL365361.1 3480 1 lncRNA 0.007184 2.48E-09 17 4.0 0 0 5.62E-09 7.30E-08
ENSG00000264920 AC018521.5 4583 17 lncRNA 0.00851 1.02E-08 16 3.7 1 0.1 2.06E-07 1.85E-06
ENSG00000231784 DBIL5P 2676 17 Transcribed_unitary_pseudogene 0.008595 6.26E-07 18 4.2 1 0.1 2.39E-08 2.63E-07
ENSG00000273033 LINC02035 5475 3 lncRNA 0.006758 9.54E-06 23 5.4 0 0.0 6.15E-12 8.61E-11
ENSG00000266979 LINC01180 3985 17 lncRNA 0.005521 1.18E-05 13 3.0 1 0.1 5.00E-06 4.00E-05
ENSG00000272070 AC005618.1 3147 5 lncRNA 0.006991 4.19E-05 18 4.2 1 0.1 2.39E-08 2.63E-07
ENSG00000251562 MALAT1 8828 11 lncRNA 0.005551 0.000283 30 7.0 19 2.2 4.19E-05 2.09E-04
ENSG00000261094 AC007066.2 2710 9 lncRNA 0.008118 0.000287 11 2.6 2 0.2 1.86E-04 7.43E-04
ENSG00000263874 LINC00672 4216 17 Protein_coding 0.005455 0.000493 14 3.3 2 0.2 9.70E-06 6.79E-05
ENSG00000180764 PIPSL 3349 10 Transcribed_processed_pseudogene 0.006569 0.0013 14 3.3 34 3.9 0.640 0.640
ENSG00000237984 PTENP1 3995 9 Transcribed_processed_pseudogene 0.005507 0.0013 15 3.5 22 2.5 0.376 0.640
ENSG00000240859 AC093627.4 5916 7 lncRNA 0.006085 0.0048 18 4.2 2 0.2 1.66E-07 1.66E-06
Fifteen noncoding genes with an enriched mutation rate compared to surrounding nonannotated regions. P-values are derived from the Fisher exact test (two-sided) and corrected for multiple testing using the FDR (Hochberg) method.
aENSG not recognised by ICGC data portal.
To further investigate the mechanisms underlying the observed SNVs and MNVs, we used the latest COSMIC mutational signatures (v3) to establish the presence and contribution of these predefined mutational signatures in metastatic CRC17. We identified 11 single base signatures (SBS) and 9 double base signatures (DBS) that had a relative contribution of at least 10% in minimally 10 cases and as such were considered dominant signatures in mCRC; SBS1, SBS5, SBS8, SBS9, SBS17b, SBS18, SBS35, SBS39, SBS40, SBS41, SBS44, DBS2-9, and DBS11. De novo signature calling using the Non-negative Matrix Factorization algorithm (NMF)18 did not identify additional signatures besides the known COSMIC signatures in our cohort.
Effects of systemic prior treatment on the genomic landscape
Patients receiving prior systemic treatment (n = 284) showed a significantly higher TMB, a higher number of SVs, a higher number of affected GISTIC CNV regions (7.58 versus 5.82; 208 versus 148; 31 versus 28, respectively; MWU p-values < 0.005), and more frequent occurrence of chromothripsis (6.5 versus 13.4%; Fisher exact test p = 0.042) compared to patients (n = 124) without prior systemic treatment. More specifically, we observed altered relative contributions for several mutational signatures in defined prior-treatment groups compared to treatment-naive patients (n = 124, Fig. 3 and Supplementary Data 2; MWU, FDR p < =5.15E-7). Patients who were prior-treated with a combination therapy of PLAT/PYR + target showed increased relative contributions of SBS8, SBS17b, SBS35, and DBS5 compared to treatment-naive patients. These results are supported by previous studies in which DBS5 and SBS35 signatures were linked to the effect of platinum (PLAT) compounds, while SBS17b was detected specifically in 5-FU or capecitabine (PYR) exposed tumors19. SBS8 was previously indirectly associated with prior platinum treatment in metastatic breast cancer17,20.Fig. 3 Mutational signatures in prior-treated cases compared to untreated cases.
Relative contribution (%) of several single and double base mutational signatures (SBS/DBS) in patients receiving prior treatment with platinum, pyrimidine antagonist, and targeted anti-EGFR treatment (PLAT/PYR + target; orange, n = 134) compared to untreated patients (blue, n = 124). Horizontal lines indicate the median. P-values are derived from the MWU test (two-sided) and corrected for multiple testing using the FDR (Hochberg) method. Source data are provided as a Source Data file.
Remarkably, even though TMB was increased in patients who received prior treatment compared to treatment-naive patients, no specific mutations (coding or noncoding) were associated with any of the defined prior-treatment groups or with prior treatment in general. With regard to the GISTIC-defined CNVs, we found increased frequencies of gains at 6p22.1, 6p21.1, and 18p11.32 as well as losses at 3p14.2 and 8p21.3 in patients who received prior treatment (Supplementary Table 2; chi-square FDR < 0.05). More specifically, gains of 6p22.1 and 6p21.1 were also associated with a prior-treatment regimen containing PLAT/PYR ± target whereas loss at 8p21.3 was only associated with PLAT/PYR + target.
Comparing metastatic CRC to primary CRC
The above described characteristics of our metastatic cohort were related to previous reports on primary CRC to identify changes potentially linked to the metastatic process (Supplementary Table 1). Therefore, we compared the observed relative contributions of the 20 dominant mutational signatures in our cohort to primary CRC data described by Alexandrov et al. (PCAWG cohort)17. For this analysis only the 124 untreated metastatic CRC cases from our cohort were included, since multiple treatments are known to specifically affect these mutational signatures17,19,20. SBS1, 8 and 41, as well as DBS2, 4, and 6 showed a significantly increased relative contribution in untreated metastatic cases (MWU, FDR ≤ 0.01; Fig. 4), suggesting they may be associated with the metastatic process. Etiologies for these signatures are either unknown (SBS8/41, DBS1) or appear age-related (SBS1, DBS2/DBS4), although DBS2 has also been linked to exposure to tobacco smoking and other endogenous and exogenous mutagens. Mutation frequencies per gene were compared between primary CRC (TCGA-DFCI cohort) and our total metastatic cohort. For this purpose, we selected genes mutated in primary CRC (TCGA-DFCI cohort) with >5% prevalence and complemented these with here identified metastatic driver genes regardless of their prevalence in primary CRC. Increased frequencies were only observed in driver genes TP53, ZFP36L2, KRAS, and APC (Fisher exact test, FDR ≤ 0.012). A decreased frequency was observed for 21 non-driver genes (Supplementary Table 3) and 1 driver gene, namely PIK3CA (Table 2). With respect to the identified putative noncoding drivers (Table 3), all of them were enriched in mCRC compared to primary CRC, except for PIPSL and PTENP1 (ICGC dataset; Fisher exact test, FDR < 5.74E-4).Fig. 4 Mutational signatures in primary CRC and untreated metastatic CRC.
Relative contribution (%) of several single and double base mutational signatures (SBS/DBS) in primary CRC tumors (purple, n = 73)17, compared to untreated metastatic CRC tumors (green, n = 124). Horizontal lines indicate the median. P-values are derived from the MWU test (two-sided) and corrected for multiple testing using the FDR (Hochberg) method. Source data are provided as a Source Data file.
Distinct mutational signature patterns in mCRC patients
Unsupervised hierarchical clustering using the 20 dominant mutational signatures complemented with mutational signatures previously described in primary CRC (SBS15/17a/28/37 and DBS10), and mutational signatures showing a dominant relative contribution (>25%) in at least one of our samples (SBS10a/10b), revealed three major and three minor groups of patients (Fig. 5).Fig. 5 Unsupervised hierarchical clustering of metastatic CRC using relative contribution of preselected mutational signatures.
Heatmap representing the median-centered relative contribution of mutational signatures between samples. Values were scaled from red (relative contribution above median) to yellow (relative contribution below median). Included single and doublet base signatures (SBS/DBS) are indicated at the right to which etiologies are added when known. Grouping of samples is shown by the dendrogram at the top. Source data are provided as a Source Data file.
The three major groups are found in cluster 1, cluster 3, and cluster 6. Clusters 1 and 6 are labeled “prior treatment” and “primary-like” as they are enriched for either patients with or without prior treatment compared to all other clusters (Fisher’s exact test: p = 4.588E-25 and p = 4.754E-15, respectively) and are characterized by higher relative contributions of signatures related to prior treatment (SBS5/8/35/17a/17b and DBS5) and signatures known from primary CRC (SBS1/5/18/40, DBS9), respectively. Samples from Cluster 6 are enriched (Fisher’s exact p = 0.005) for samples with >5% contribution of the recently described E. coli mutational signature in CRC as well21. Cluster 3 was labeled ‘mCRC-specific’ as it contains both patients with (n = 63) and without (n = 31) prior treatment characterized by higher relative contributions of signatures SBS9/37/39/41, which, except for SBS37, are rarely detected in primary CRC. Etiologies for SBS37/39/41 are unknown, whereas SBS9 mutations have been partly associated with polymerase eta (Pol η) function during somatic hypermutation in lymphoid cells. In vitro, Pol η activity has been associated with anticancer drugs resistance, specifically cisplatin and 5-FU22–24. Indeed we find that the majority of patients (13 out of 15) in cluster 3 with a high SBS9 contribution (≥10%) had already received prior treatment, although this did not reach statistical significance (Fisher’s exact test p = 0.07).
The remaining minor groups are found in Clusters 2, 4, and 5. Samples in clusters 2 and 4 are defined by a large contribution of DBS8 and DBS2, respectively. Cluster 5, labeled ‘high TMB’, contains 14 samples, which were all characterized by a high TMB (defined as >10/Mb) compared to only 82 out of the 415 remaining samples (20%) in the other clusters. High contributions of DNA mismatch repair associated signatures SBS15/44 and DBS7 characterize the 13 MSI samples in this cluster, whereas the one remaining sample showed high contributions of SBS10a/b, associated with polymerase epsilon (POLE) mutations.
MSI-specific gene mutations
We subsequently investigated whether specific somatic gene mutations were associated with each of the six clusters described above and found this was true only for the high TMB cluster (cluster 5). To correct for the higher likelihood of finding any mutation in a high TMB sample, we applied a permutation test25,26, which identified 28 genes as significantly more frequently mutated in the high TMB cluster versus all other samples (Fisher exact test, FDR and permutation p < 0.05, see Supplementary Table 4). As these 28 genes are large (cDNA size range 1.5–22 kb) and often contain substantial numbers of microsatellites and mononucleotide stretches (range 4–126), we evaluated whether their observed mutation frequency in MSI cases was significantly higher than the frequency distribution observed for all other genes with a comparable number (±10%) of MSI-prone coding sequences. Except for TNXB, for which we were unable to establish a reliable control distribution, all identified genes were significantly more frequently mutated in MSI cases compared to control genes containing similar numbers of MSI-prone sequences (one sample sign test; all p ≤ 0.0001). These results suggest that mutations in these genes are selected for during the disease process in MSI tumors. The top 2 genes, ACVR2A and UBR5, are known targets of the MSI process27. LRP1 mutations were found to reduce its expression in CRC and were associated with MSI status and poor outcome28. Although the other 25 identified genes were not previously associated with MSI status, three of these genes (KMT2C, KMT2D, and FAT1) were present in the Yaeger dataset of mCRC samples6. Mutations in all three overlapping genes were significantly enriched in MSI cases (n = 16) compared to microsatellite stable (MSS) cases (n = 305) in this dataset as well (all Fisher p < =9.19E-7).
Association between molecular landscape and treatment response
The observed molecular characteristics were associated with response to current treatment for the 286 patients in our cohort with recorded treatment response. These results should be interpreted with caution due to the heterogeneity of our cohort in terms of both treatment line and type of prior treatments received, which may introduce bias. We studied ordinal response (PD, SD, and PR) to any treatment as well as to specific treatment regimens. In total, 123 items were used as input in the regression model, consisting of five themes (full list in Supplementary Data 3): clinical parameters (age, gender, prior treatment, and radiotherapy), counts (TMB, kataegis, chromothripsis, total number of SV by type and the number of 10kb–1Mb deletions), mutational signatures (DBS/SBS), driver genes (including noncoding genes), and GISTIC-defined CNVs. Items that reached univariate statistical significance (p < 0.05) were used in a multivariable penalized ordinal regression model for treatment response (Table 4).Table 4 Multivariate LASSO analysis.
Type Item All treatments Oxaliplatin containing PLAT/PYR Target-mono TOP/PYR
Clinical Prior Treatment 0.57
Gender 0.77 −1.14 −1.10
Counts nr of Tandem Duplications 1.19
nr of 10kb–1Mb deletions 0.01
Mutational Signatures DBS2 −0.02
DBS5 0.13
DBS11 −0.03
SBS17b 0.07 0.13
SBS39 0.04
SBS41 −0.21
Driver Genes APC 0.23
KRAS 0.78
PIK3CA 0.22
FBXW7 17.65
Non-coding LINC00672 0.90
GISTIC Regions Gain 17q12 (ERBB2*) 0.44
Gain 18p11.32 (CETN1*) 0.59
Gain 20q11.1 (BCL2L1*) 0.12
Gain 8p11.21 (KAT6A) −0.78 −1.56
Gain 7p21.3 (VWDE) 3.68
Gain 7q31.2 (MET*) 3.59
Gain 7p12.3 (PKD1L1) 3.04
Gain 7q34 (no genes in peak) 3.59
Gain 14q23.1 (no genes in peak) 1.64
Loss 18q12.2 (hsa-mir-924*) −1.52 −3.30 2.79
Loss 6q26 (PARK2) −1.38
Loss 9p21.3 (CDKN2A*) −1.88
Loss 16q23.1 (WWOX) −1.70
Loss 4q22.1 (CCSER1) 2.66
Loss 4q35.1 (IRF2) 1.78
Loss 18q21.2 (SMAD4) −1.35 −3.30 2.79
Loss 18q23 (NFATC1*) −1.14 −3.30
Loss Xp22.31 (STS*) −1.40 1.78
Loss 14q23.3 (GPHN) −2.09
Items that reached univariate statistical significance (p < 0.05) were used in a multivariable penalized ordinal regression model for treatment response. Univariate regression was performed for genomic features (Supplementary Data 3) using the ‘polr’ function from the MASS R package (v7.3-51.4) and subsequently those with a univariate two-sided p-value <0.05 were selected for multivariable ordered LASSO regression using the ordinalNet R package (v2.7). Regression coefficients are shown for features that remained significant in the multivariable model.
*Multiple genes present in region; bold: known Fragile Site region
Overall we found that, next to receiving prior treatment(s), the number of 10kb–1Mb deletions, mutations in KRAS, APC, PIK3CA, and LINC00672, mutational signatures SBS17b/39, DBS2/5/11, and gains at 18p, 17q, and 20q were associated with treatment response regardless of treatment type in mCRC patients. For SBS17b this effect was more pronounced when specifically investigating patients treated with platinum as described before17. CNVs were predominantly associated with response to PLAT/PYR or PYRmono treatment, whereas mutations in FBXW7 were associated with poor response to targeted treatment. FBXW7 mutations were detected in 51 patients from our cohort, including 21 KRAS wild-type patients. Of these 21 patients, five were treated with panitumumab monotherapy, all of whom had PD as best response. This suggests that, next to somatic KRAS mutations, somatic FBXW7 mutations may provide an additional negative selection marker for anti-EGFR treatment. This finding is in concordance with previous reports on FBXW7 mutation prevalence in nonresponding patients on anti-EGFR treatment29,30.
Potential clinical implications
WGS data of our cohort of 429 patients with metastatic CRC revealed several potential molecular features that might be associated with sensitivity to particular anticancer agents. A high TMB (here defined as >10 mutations per Mb) has been suggested as a potential selection tool for tumors that may respond to immunotherapy31. In our cohort, 96 (22%) samples showed a TMB > 10, of which 13 were MSI. A gradual increase in TMB was observed with the number of prior treatments (test for trend, p = 4.39E-13). For the subset of samples of which we also had RNA-seq data available, we calculated the Tumor Infiltrating Leukocyte (TIL) score as a proxy for the immunogenicity of the tumor32. Interestingly, we did not observe a significantly higher TIL score in the TMB-high samples (n = 21) compared to the other samples (n = 63; MWU; p = 0.39), whereas the average TIL score in MSI samples is significantly higher compared to both MSS samples with a high TMB and with a low TMB (Kruskal–Wallis test (p = 0.037) followed by Dunn’s pairwise comparison (Benjamini–Hochberg corrected p = 0.012 and p = 0.021 for MSI compared to MSS with high and low TMB, respectively (See Supplementary Fig. 2). Although far from definite, these results support the on-label use of immunotherapy in MSI tumors and suggest that merely using TMB may not be sufficient to identify the tumors with immunogenic potential in the metastatic setting.
Other on-label markers found in our cohort include a targetable BRAF V600E mutation in 40 patients, as well as 130 RAS/RAF wild-type patients that did not receive targeted anti-EGFR treatment yet. However, our data suggest that mutations in FBXW7, observed in 21 out of these 130 RAS/RAF wild-type patients, should be considered as a contra-indication for the use of anti-EGFR treatment. Molecular biomarkers for potential off-label use that were found in our cohort include amplifications of ERBB2 (HER2), MET and CDK4, loss of BRCA1 and BRCA2 through deletion or high impact mutations, loss of TSC1 and TSC2 through high impact mutations, and possible fusions of PDGFRB. In addition, 23 patients in our cohort carried a KRAS G12C mutation, for which an inhibitor may become available in the near future33.
In summary, for 55% of our patients one or more targeted treatments are potentially available based on the molecular profile of their cancer (Fig. 6).Fig. 6 Actionable genes.
Data from OncoKB were matched to affected genes observed in our mCRC cohort. Numbers indicate the number (and percentage) of affected patients. Source data are provided as a Source Data file.
Discussion
This study encompasses a WGS-based, comprehensive description of the molecular landscape of metastatic CRC and aims to put this landscape into perspective by associating it with prior systemic treatments, comparing it to primary CRC and relating it to treatment response.
In general, the genomic landscape of CRC remains relatively stable in metastatic disease. However, compared to primary CRC, our metastatic CRC cohort showed significant enrichment for mutations in 4 out of 23 coding and 12 out of 15 noncoding (putative) driver genes. From the identified putative drivers, only mutations in PIK3CA were significantly decreased in mCRC. Six of our identified coding driver genes are not present in the current CRC-specific MSK-IMPACT panel, namely ZFP36L2, BCL, BCL9L, ELF3, LMTK3, and TGIF1.
Within the mCRC cohort we observed clear effects of received prior treatments on the total numbers of aberrations, CNVs, and mutational signatures, with the latter sufficiently dominant to show up as a separate group after hierarchical clustering. Remarkably, we also observed an mCRC-specific cluster characterized by signatures which are rarely found in primary CRC and are not associated with any treatment (SBS9/39/41). SBS9 is associated with Pol η activity, an error-prone polymerase encoded by the POLH gene, which mediates translesion synthesis and is induced by replication stress34. Interestingly, high levels of Pol η have been associated with cancer therapy resistance in vitro22–24. We did observe that the majority of patients with a high relative SBS9 contribution had already received prior treatment; however, unfortunately, sample numbers were too low to directly associate SBS9 contribution with POLH expression in our dataset. Another predominant cluster group consisted of metastatic MSI samples. In these samples we observed a significant enrichment of mutations in a specific set of genes compared to other similarly MSI-prone genes, suggesting these genes are preferentially affected or selected for during disease progression.
The varying number and types of received prior treatments within our cohort hampered the search for prognostic and predictive biomarkers. However, we found that, next to already known events, the number of LINC00672 mutations and 10kb–1Mb deletions were associated with treatment response irrespective of the type of treatment. Strikingly, many of these recurrent deletions occur in known Common Fragile Site (CFS) genes, as described in primary CRC as well13, implicating replication stress as one of driving mechanisms35. In addition, FBXW7 mutations were predictive for poor response to EGFR-targeted treatments in our prospective cohort. This is in line with previous observations showing that FBXW7 mutations were enriched in unresponsive patients compared to patients responding well to EGFR-targeted treatments29,30.
The current study gives a detailed description of the genomic landscape of metastatic CRC. More specifically, our study identifies treatment-induced changes, metastasis-specific alterations, and associations between molecular traits and response to treatment. In addition, we provide prospective validation for FBXW7 mutations as a predictive biomarker for poor response to EGFR-targeted treatment. Combined with future studies, this catalogue of molecular alterations will speed up the identification of resistance mechanisms, the determination of metastasis-driving processes, and, ultimately, the improvement of metastatic CRC patient care.
Methods
Patient cohort and study procedures
Colorectal cancer patients included in this study were selected from the previously described cohort of the Center for Personalized Cancer Treatment (CPCT) consortium (CPCT-02 Biopsy Protocol, ClinicalTrial.gov no. NCT01855477), which was approved by the medical ethics committee of the University Medical Center Utrecht, the Netherlands9. All patients have given explicit consent for whole-genome sequencing and data sharing for cancer research purposes. Upon our data request for all CRC patients thus far, we were provided with the data of all patients registered as metastatic CRC patients included between April 2016 and January 2019 (n = 487). Patients who received systemic treatment which is not normally given to colorectal cancer patients (e.g., carboplatin, paclitaxel, sunitinib, and etoposide) were excluded to avoid erroneous inclusion of patients suffering from another type of cancer (n = 28). When multiple biopsies were included for one patient (n = 29), only the first biopsy was included in our analyses. In total, we included 429 distinct CRC patients in our analyses. Based on the provided information regarding all forms of systemic treatment patients received before the study biopsy took place (further referred to as “prior treatment”), we coded the (groups of) active agents using the following abbreviations: PLAT (oxaliplatin), PYR (fluoropyrimidines), TOP (topoisomerase inhibitor; Irinotecan), +targeted (when bevacizumab or panitumumab was added), CHEMCOM (triplet combination therapy). Prior-treatment regimens were grouped based on their working mechanism to enable the analysis of their effect on the genomic landscape. Treatment related analyses were performed using combinations of the abbreviations mentioned above. For detailed information see Table 1.
Whole-genome sequencing; identification of somatic changes
Whole-genome sequencing of paired tumor/normal was performed in all cases. In short, raw sequencing data were processed using bcl2fastq (versions 2.17 to 2.20), mapped to the human reference genome GRCh37 using BWA-mem v0.7.Sa and GATK BQSR and Haplotype Caller v3.4.46 and Strelka v1.0.14 were used to call somatic mutations. Within our cohort, 98% of the biopsies of metastatic lesions showed a coverage of at least 30× (95% with >60× coverage), whereas for the normal blood 98% had >10× coverage and 94% >20× coverage. The identification of copy number changes was performed using GISTIC v2.0.2336 with the following parameters: genegistic 1; gcm extreme; maxseg 4000; broad 1; brlen 0.98; conf 0.95; rx 0; cap 3; saveseg 0; armpeel 1; smallmem 0; res 0.01; ta 0.1; td 0.1; savedata 0; savegene 1; and qvt 0.19,20.
RNA sequencing and CMS calling
Matched RNA was isolated from the same frozen tissue for 91 CRC patients on an automated setup (QiaSymphony) according to supplier’s protocols (Qiagen) using the QIAsymphony RNA Kit for tissue and quantified by Qubit. A total of 50–100 ng of RNA was used as input for library preparation using the KAPA RNA HyperPrep Kit with RiboErase (Human/Mouse/Rat) (Roche). Barcoded libraries were equimolarly pooled and sequenced using standard settings (Illumina) on either a NextSeq 500 (V2.5 reagents) generating 2 × 75 read pairs or a NovaSeq 6000 generating 2 × 150 read pairs. BCL output was converted to FASTQ using bcl2fastq (versions 2.17–2.20) using default parameters and sequence reads were trimmed for adapter sequences using fastp (v0.20.0). The resulting FASTQ files were mapped to GRCh38 using STAR (v2.6.1d)37. Sambamba (v0.7.0)38 was used to mark duplicates and index the resulting BAM files. Gene annotation was derived from GENCODE Release 30 (https://www.gencodegenes.org/), raw read counts were obtained with featureCounts (v1.6.3)39 and normalized using GeTMM40. Normalized data were used to (1) determine CMS with both the single-sample prediction parameter from the “CMSclassifier” package (v1.0.0) (https://github.com/Sage-Bionetworks/CMSclassifier)2 and CMSCaller v(0.99.1)11, and (2) calculate the Tumor Infiltrating Lymphocytes (TIL) score by averaging the expression of TIL-genes32
Identification of mutational signatures and driver genes
Mutational signatures (COSMIC v3)17 were called using R package MutationalPatterns v1.10.018, focusing on single and double base signatures. This package was also used to perform de novo signature calling using the Non-negative Matrix Factorization (NMF) method. Detection of kataegis and chromothripsis was performed as previously described41. In short, to call kataegis only SNVs were considered to establish segments based on the intermutational distance. Segments were determined using a piecewise constant fitting model and were called as kataegis when at least five SNVs were present showing an intermutational distance of ≤2 kb. Chromothripsis-like events were called using the Shatterseek R package (v0.4). Driver genes, i.e., genes under selective pressure, were identified by the dN/dS model using R package dndscv (v0.0.1.0)42. A global q ≤ 0.05 was used to select statistically significant driver genes. The R package discover v0.9.243 was used to test for mutual exclusivity. To identify noncoding genes with an enriched mutation rate, we first established a baseline mutation rate based on all identified SNVs, MNVs and Indels found in nonannotated regions, as we assume these regions are not under any selective pressure. Nonannotated regions were based on GENCODE annotation (version33) and for each of these regions we calculated a mutation rate (number of mutations/size of region). Next, a mutation rate (number of mutations/size of noncoding gene) was calculated for all somatic mutations annotated as ‘noncoding transcript variant’. The signed-rank test with Hochberg’s multiple testing correction was used to determine whether the mutation rate of a recurrent noncoding gene (mutated in at least 5% of the cohort) exceeded the baseline mutation rate. Per noncoding gene a specific baseline was determined using only nonannotated regions (>1 kb) in an area of 2 Mb surrounding the respective noncoding gene.
Verification in publicly available datasets
To compare mutational signatures, publicly available WGS data from 73 primary colorectal cases were used17. We downloaded the matrix of counts for single and double base substitutions of primary cases and analyzed these in the same manner as the metastatic CRC cases to call mutational signatures. Observed frequencies of mutated genes in metastatic CRC were verified and compared in two publicly available datasets. Dataset 1, the Yaeger dataset, contained 321 unique metastatic CRC patients that were profiled for mutations by targeted sequencing6. Dataset 2, the TCGA-DFCI dataset, contained 1949 unique primary CRC patients that were profiled for mutations in coding regions (accessed via cBioPortal January 21, 2020). Prior to analysis, synonymous mutations were removed and multiple mutations within the same gene were aggregated per patient. Dataset 3, the ICGC dataset, was used to compare mutation frequencies of noncoding genes and contained 866 unique primary CRC patients with available mutation data (accessed via the ICGC data portal, release 28). The used cohorts are summarized in Supplementary Table 1.
Estimating MSI-prone sequences
To evaluate preferentially mutated genes in MSI cases, the number of MSI-prone sequences in a gene are of interest. Data of the Microsatellite Database (MSDB, https://data.ccmb.res.in/msdb/, June 2, 2020) were filtered for repeats annotated to human exons44. For each gene, the number of repeats was summed. In addition, a custom Perl script was used to count mononucleotide stretches of lengths between 6 and 13 (the latter is the minimum length used in MSDB) as we noticed many InDels in our data in mononucleotide stretches less than 13 bases long. Exon sequences of the Consensus CDS database (https://www.ncbi.nlm.nih.gov/CCDS/) were used to count the number of mononucleotide stretches per gene.
Associations with response to treatment
Treatment response was evaluated according to RECIST (v.1.1) every 8 to 12 weeks depending on the treatment regimen and was defined as response (partial or complete), stable disease, or progressive disease45. For regression analyses, the best overall response was used as outcome measure. Genomic features (at least 5 events per group) were associated with response to treatment in a 2-step procedure using ordinal LASSO (least absolute shrinkage and selection operator) regression, which is suited for datasets with a relatively high number of predictors in comparison to cases and protects against overfitting. First, univariate regression was performed for genomic features (Supplementary Data 3) using the ‘polr’ function from the MASS R package (v7.3–51.4) and subsequently those with a univariate p-value <0.05 were selected for multivariable ordered LASSO regression using the ordinalNet R package (v2.7).
Identification of potentially actionable events
OncoKB (accessed on March 31, 2020) was used to identify clinically actionable genes from the list of mutated genes in our cohort, using only genes with level 1 and 2 evidence46. In case OncoKB listed a specific gene alteration as actionable genomic aberration, we only counted patients that harbored that specific mutation or CNV. For genes for which only ‘Oncogenic mutations’ were listed by OncoKB, we only included patients if the gene had a mutation with ‘High impact’ consequence (i.e., a nonsense or frameshift mutation). To evaluate patients eligible for an anti-EGFR therapy, we included only patients that were triple wild-type for KRAS, NRAS, and BRAF, and excluded those patients that had already received anti-EGFR therapy prior to biopsy.
Statistics
In general, a Pearson’s Chi-squared test or Fisher’s exact test (in case of too few expected events) was used to evaluate the categorical data while continuous variables were evaluated using either a Mann–Whitney U test (MWU) or a Kruskal–Wallis H (KWH) test depending on the number of categories. All statistical tests were two-sided and considered statistically significant when P < 0.05. Stata 13.0 (StataCorp) and R (v3.6.0) were used for the statistical analyses. Multiple testing using the Hochberg procedure to correct P values was applied when necessary. The statistical test used is specified throughout the results section.
Reporting summary
Further information on research design is available in the Nature Research Reporting Summary linked to this article.
Supplementary information
Supplementary Information
Peer Review File
Description of Additional Supplementary Files
Supplementary Data 1
Supplementary Data 2
Supplementary Data 3
Reporting Summary
Source data
Source Data
Supplementary information
Supplementary information is available for this paper at 10.1038/s41467-020-20887-6.
Acknowledgements
We thank the Hartwig Medical Foundation, and Stichting Stelvio for Life for financial support of clinical studies and WGS analyses. We thank the Center for Personalized Cancer Treatment for proving the clinical data. We would like to thank all local principal investigators, medical specialists, and nurses of all contributing centers for their help with patient accrual. We are particularly grateful to all participating patients and their families.
Author contributions
P.A.J.M., M.S., S.S., J.W.M.M., and S.M.W. wrote the manuscript, which all authors reviewed. M.S. and J.V.R. performed the bioinformatics analyses. P.A.J.M., L.A., and S.S. managed clinical data assessment. M.L., N.S., M.P.H., G.A.C., J.M.V.R., and A.J.T.T. are main clinical contributors. M.P.L. and S.S. are members of the CPCT-02 study team and/or CPCT board. E.C. coordinated the sequencing of samples and contributed to the bioinformatics analyses.
Data availability
The WGS, RNA-seq, and corresponding clinical data used in this study was made available by the Hartwig Medical Foundation (Dutch nonprofit biobank organization) after signing a license agreement stating data cannot be made publicly available via third party organizations. Therefore, the data are available under restricted access and can be requested upon by contacting the Hartwig Medical Foundation (https://www.hartwigmedicalfoundation.nl/applying-for-data/) under the accession code DR-058. Publicly available datasets that were used in this study are listed in Supplementary Table 1. The Yaeger data used in this study are available in the cBioPortal for Cancer Genomics (http://www.cbioportal.org/study?id=crc_msk_2017). The TCGA-DFCI data used in this study have been deposited in the cBioPortal for Cancer Genomics which we accessed on January 21, 2020 (https://www.cbioportal.org/study/summary?id=coadread_tcga; https://www.cbioportal.org/study/summary?id=coadread_tcga_pub; https://www.cbioportal.org/study/summary?id=coadread_tcga_pan_can_atlas_2018; https://www.cbioportal.org/study/summary?id=coadread_dfci_2016). The ICGC data used in this study have been deposited in the ICGC data portal (release 28) (https://dcc.icgc.org/projects/COAD-US; https://dcc.icgc.org/projects/COCA-CN; https://dcc.icgc.org/projects/READ-US). The remaining data are available within the Article, Supplementary Information or available from the authors upon request. Source data are provided with this paper.
Code availability
The bioinformatical code used for data processing is available at https://github.com/hartwigmedical/pipeline5.
Competing interests
The authors declare no competing interests.
Peer review information Nature Communications thanks Kin Chan, Silvia Marsoni and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
These authors contributed equally: Pauline A.J. Mendelaar, Marcel Smid.
Change history
5/27/2021
A Correction to this paper has been published: 10.1038/s41467-021-23629-4 | PANITUMUMAB | DrugsGivenReaction | CC BY | 33495476 | 18,948,086 | 2021-01-25 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Kaposi^s sarcoma'. | Extensive cutaneous iatrogenic Kaposi's sarcoma after bullous pemphigoid treatment with oral methylprednisolone: a rare Chinese case report.
Bullous pemphigoid (BP) is an autoimmune disease that requires immunosuppressive therapy. Systemic corticosteroids are considered the standard treatment for moderate-to-severe BP. Kaposi's sarcoma (KS) is a rare multifocal endothelial tumour that affects the skin, mucosa and viscera. As an angioproliferative disease of obscure aetiopathogenesis and histogenesis, KS is associated with human herpesvirus 8 (HHV-8). This current case report describes a rare occurrence of extensive cutaneous KS in a 60-year-old Chinese male patient after oral methylprednisolone treatment for BP with an emphasis on its pathological characterization. A total of more than 40 nodules were found on his trunk and lower limbs covering more than 20% of his body surface area. Immunohistochemical staining of biopsy samples from the lesion showed the patient was positive for HHV-8, CD31, CD34, XIIIa, ERG and Ki-67. The Epstein-Barr virus test showed the patient tested negative for immunoglobulin (Ig)A and IgM, but was positive for IgG. Immunosuppression associated with the treatment for BP may activate a latent HHV-8 infection and induce the development of KS.
Introduction
Kaposi’s sarcoma (KS) is a rare multifocal endothelial tumour that affects the skin,
mucosa and viscera.1 As an angioproliferative disease with obscure aetiopathogenesis and
histogenesis, KS is associated with human herpesvirus 8 (HHV-8).2 KS is mainly diagnosed in elderly Ashkenazi Jewish or Mediterranean
men3,4 and is extremely
rare in China.5 Immunocompromised patients, including those undergoing immunosuppressive
therapy for bullous diseases, have a higher risk for developing KS.6 KS is classified into four types based on the clinical circumstances in which
it develops: classic (originally described by Kaposi, which typically presents in
middle or old age); endemic (described in young adult males in sub-Saharan
indigenous Africans and can be more aggressive); iatrogenic (associated with
immunosuppressive drug therapy); and epidemic KS (acquired immunodeficiency syndrome-associated).7 This case report describes a rare occurrence of KS in a male patient treated
with methylprednisolone for generalized bullous pemphigoid (BP).
Case report
A 60-year-old male patient visited the outpatient department at Huashan Hospital,
Fudan University, Shanghai, China in March 2018 with a 10-month history of
generalized BP (Figure 1).
He was treated with methylprednisolone for 7 months (highest dose 60 mg/day orally
once daily for 2 weeks, gradually decreased to 22 mg/day orally once daily for 3
weeks). In October 2018, the patient developed erythaema and nodules on the trunk
and all extremities. His treatment with 22 mg/day methylprednisolone orally once
daily was continued. A history of drug addiction, alcohol consumption and smoking
was denied. Physical examination revealed multiple (more than 40), discrete,
hyperkeratotic, excoriated papules and nodules (∼0.5–2 cm in diameter). These were
on his trunk (about 10), neck and the extensor surfaces of his limbs (lower limbs
about 30) covering about 20% of his body surface area (Figures 2a–2c). No other abnormalities were
found on clinical examination, including chest and abdominal computed tomography
imaging. Laboratory tests revealed that the patient tested negative for human
immunodeficiency virus (HIV), the rapid plasma reagin test and treponema pallidum
particle agglutination assay. Other routine tests revealed normal findings. There
were no signs of a tumour. A biopsy was taken from a lesion on his right lower limb
and the tissue specimen was stained with haematoxylin and eosin (Figures 2d–2f) and subjected
to immunohistochemical analysis (Figure 3) for CD31(+), CD34(+), HHV-8(+), XIIIa(+), ERG(+) and Ki-67(+).
The Epstein–Barr virus test showed that the patient tested negative for
immunoglobulin (Ig)A and IgM, but was positive for IgG. The patient was diagnosed
with iatrogenic KS and BP. No treatment was administered because the patient
requested to be discharged and went back to his local hospital. The patient was lost
to follow-up.
Figure 1. The clinical features of bullous pemphigoid on the forearms, trunk and legs
of a 60-year-old male patient during remission (a, b and c). Haematoxylin
and eosin staining of a skin biopsy confirmed a diagnosis of bullous
pemphigoid (d); scale bar 50 µm. The colour version of this figure is
available at: http://imr.sagepub.com.
Figure 2. Multiple discrete papules and nodules were observed on the trunk and limbs of
a 60-year-old male patient. Some bullae developed on normal-appearing skin
or papular/nodular lesions (a–c). Haematoxylin and eosin staining of a
biopsy of a Kaposi’s sarcoma lesion showed hyperkeratosis, acanthosis,
irregular elongation of rete ridges and subepidermal cavities; scale bar
100 µm (d); scale bar 50 µm (e); and scale bar 10 µm (f). The colour version
of this figure is available at: http://imr.sagepub.com.
Figure 3. Immunohistochemical analysis of a biopsy of a Kaposi’s sarcoma lesion for
CD31, CD34, human herpesvirus 8 (HHV-8), XIIIa, ERG and Ki-67. Scale bar
50 µm. The colour version of this figure is available at: http://imr.sagepub.com.
Written informed consent was obtained from the patient’s guardians for publication of
this case report and the accompanying images. The study was approved for publication
by the Ethics Committee of Huashan Hospital (no. KY2018-457). Written Informed
consent was obtained from the patient for treatment.
Discussion
Bullous pemphigoid is an autoimmune bullous disease that requires immunosuppressive therapy.8 The pathogenesis of BP is attributed to autoantibodies directed against
distinct adhesion molecules of the epidermis or the dermoepidermal junction.9 Patients with BP usually present with large, fluid-filled blisters on
flexural areas of the skin.9 Pharmacotherapy for BP, which includes corticosteroid medications such as
prednisone and other drugs that suppress the immune system, helps heal the blisters
and alleviates itching.9 However, BP can be life-threatening, especially for elderly individuals in
poor health.10 Only a few cases of KS associated with immunosuppressive therapy for
dermatological diseases have been reported in the literature.11–18 Due to the low incidence of
these diseases, there is a lack of retrospective studies in the Chinese population
on the frequency of KS in the patients with bullous diseases. Interestingly, the
occurrence of classic KS in the Xinjiang Uygur Autonomous Region, the northwest part
of China, is high probably due to the population being composed of 59.43% minority
ethnic groups.5 According to HLA-B gene analysis, some of these people
originated in Europe 3000 years ago, migrated to Central Asia and Xinjiang, where
they then merged with the local ethnic groups and evolved into the current
race.19,20
Kaposi’s sarcoma is caused by HHV-8 infection, which is considered to be a triggering
factor for bullous diseases, especially pemphigus.21,22 However, the association
between HHV-8 and BP remains controversial.23 A recent retrospective study in non-HIV KS patients reported that 14 (1.03%)
of 1362 patients with classic or iatrogenic KS also suffered from bullous disease.24 Half of these 14 patients developed KS after the onset of bullous disease,
whereas the other half developed a bullous disease after being diagnosed with KS.24 Only a few cases on concurrent HHV-8 and BP have been reported in the
literature.12,21,22 HHV-8 identification in KS with BP has been reported only in
one patient in the Chinese population.25
There are discrepancies pertaining to HHV-8 positivity in patients with KS-associated
BP.12,14 In 2001, a
report of a case of KS after immunosuppressive therapy for BP demonstrated that
serum HHV-8 protein was detectable 4 months before the appearance of KS and
initiation of prednisolone therapy.11 The findings in this previous case suggested that the immunosuppression
associated with the treatment for BP possibly activated a latent HHV-8 infection and
induced the development of KS.11 The current patient tested positive for HHV-8 on a biopsy of the lesion. The
experience with this current patient suggests that HHV-8 positivity may be an
indicator for potential development of KS in BP patients. However, it is not clear
whether HHV-8 induced BP and subsequently KS. To clarify the sequence and causation,
it is necessary to compare the incidence of HHV-8-positive and HHV-8-negative KS
appearing after the onset of BP.
Prednisolone and azathioprine are the drugs most frequently implicated in
immunosuppression-associated KS.15 It is difficult to evaluate the relationship between the type, dosage and
duration of immunosuppressive medication and the onset of KS. The mean time to KS
onset, according to the current literature, is 4 months from the initiation
immunosuppressive therapy.26 In this current patient, BP treatment had been administered for 7 months
before the onset of KS.
Declaration of conflicting interest: The authors declare that there are no conflicts of interest.
Funding: This research received no specific grant from any funding agency in the public,
commercial, or not-for-profit sectors.
ORCID iD: Xiao-Qun Luo https://orcid.org/0000-0002-5345-6517 | METHYLPREDNISOLONE | DrugsGivenReaction | CC BY-NC | 33496629 | 19,437,819 | 2021-01 |
What was the administration route of drug 'METHYLPREDNISOLONE'? | Extensive cutaneous iatrogenic Kaposi's sarcoma after bullous pemphigoid treatment with oral methylprednisolone: a rare Chinese case report.
Bullous pemphigoid (BP) is an autoimmune disease that requires immunosuppressive therapy. Systemic corticosteroids are considered the standard treatment for moderate-to-severe BP. Kaposi's sarcoma (KS) is a rare multifocal endothelial tumour that affects the skin, mucosa and viscera. As an angioproliferative disease of obscure aetiopathogenesis and histogenesis, KS is associated with human herpesvirus 8 (HHV-8). This current case report describes a rare occurrence of extensive cutaneous KS in a 60-year-old Chinese male patient after oral methylprednisolone treatment for BP with an emphasis on its pathological characterization. A total of more than 40 nodules were found on his trunk and lower limbs covering more than 20% of his body surface area. Immunohistochemical staining of biopsy samples from the lesion showed the patient was positive for HHV-8, CD31, CD34, XIIIa, ERG and Ki-67. The Epstein-Barr virus test showed the patient tested negative for immunoglobulin (Ig)A and IgM, but was positive for IgG. Immunosuppression associated with the treatment for BP may activate a latent HHV-8 infection and induce the development of KS.
Introduction
Kaposi’s sarcoma (KS) is a rare multifocal endothelial tumour that affects the skin,
mucosa and viscera.1 As an angioproliferative disease with obscure aetiopathogenesis and
histogenesis, KS is associated with human herpesvirus 8 (HHV-8).2 KS is mainly diagnosed in elderly Ashkenazi Jewish or Mediterranean
men3,4 and is extremely
rare in China.5 Immunocompromised patients, including those undergoing immunosuppressive
therapy for bullous diseases, have a higher risk for developing KS.6 KS is classified into four types based on the clinical circumstances in which
it develops: classic (originally described by Kaposi, which typically presents in
middle or old age); endemic (described in young adult males in sub-Saharan
indigenous Africans and can be more aggressive); iatrogenic (associated with
immunosuppressive drug therapy); and epidemic KS (acquired immunodeficiency syndrome-associated).7 This case report describes a rare occurrence of KS in a male patient treated
with methylprednisolone for generalized bullous pemphigoid (BP).
Case report
A 60-year-old male patient visited the outpatient department at Huashan Hospital,
Fudan University, Shanghai, China in March 2018 with a 10-month history of
generalized BP (Figure 1).
He was treated with methylprednisolone for 7 months (highest dose 60 mg/day orally
once daily for 2 weeks, gradually decreased to 22 mg/day orally once daily for 3
weeks). In October 2018, the patient developed erythaema and nodules on the trunk
and all extremities. His treatment with 22 mg/day methylprednisolone orally once
daily was continued. A history of drug addiction, alcohol consumption and smoking
was denied. Physical examination revealed multiple (more than 40), discrete,
hyperkeratotic, excoriated papules and nodules (∼0.5–2 cm in diameter). These were
on his trunk (about 10), neck and the extensor surfaces of his limbs (lower limbs
about 30) covering about 20% of his body surface area (Figures 2a–2c). No other abnormalities were
found on clinical examination, including chest and abdominal computed tomography
imaging. Laboratory tests revealed that the patient tested negative for human
immunodeficiency virus (HIV), the rapid plasma reagin test and treponema pallidum
particle agglutination assay. Other routine tests revealed normal findings. There
were no signs of a tumour. A biopsy was taken from a lesion on his right lower limb
and the tissue specimen was stained with haematoxylin and eosin (Figures 2d–2f) and subjected
to immunohistochemical analysis (Figure 3) for CD31(+), CD34(+), HHV-8(+), XIIIa(+), ERG(+) and Ki-67(+).
The Epstein–Barr virus test showed that the patient tested negative for
immunoglobulin (Ig)A and IgM, but was positive for IgG. The patient was diagnosed
with iatrogenic KS and BP. No treatment was administered because the patient
requested to be discharged and went back to his local hospital. The patient was lost
to follow-up.
Figure 1. The clinical features of bullous pemphigoid on the forearms, trunk and legs
of a 60-year-old male patient during remission (a, b and c). Haematoxylin
and eosin staining of a skin biopsy confirmed a diagnosis of bullous
pemphigoid (d); scale bar 50 µm. The colour version of this figure is
available at: http://imr.sagepub.com.
Figure 2. Multiple discrete papules and nodules were observed on the trunk and limbs of
a 60-year-old male patient. Some bullae developed on normal-appearing skin
or papular/nodular lesions (a–c). Haematoxylin and eosin staining of a
biopsy of a Kaposi’s sarcoma lesion showed hyperkeratosis, acanthosis,
irregular elongation of rete ridges and subepidermal cavities; scale bar
100 µm (d); scale bar 50 µm (e); and scale bar 10 µm (f). The colour version
of this figure is available at: http://imr.sagepub.com.
Figure 3. Immunohistochemical analysis of a biopsy of a Kaposi’s sarcoma lesion for
CD31, CD34, human herpesvirus 8 (HHV-8), XIIIa, ERG and Ki-67. Scale bar
50 µm. The colour version of this figure is available at: http://imr.sagepub.com.
Written informed consent was obtained from the patient’s guardians for publication of
this case report and the accompanying images. The study was approved for publication
by the Ethics Committee of Huashan Hospital (no. KY2018-457). Written Informed
consent was obtained from the patient for treatment.
Discussion
Bullous pemphigoid is an autoimmune bullous disease that requires immunosuppressive therapy.8 The pathogenesis of BP is attributed to autoantibodies directed against
distinct adhesion molecules of the epidermis or the dermoepidermal junction.9 Patients with BP usually present with large, fluid-filled blisters on
flexural areas of the skin.9 Pharmacotherapy for BP, which includes corticosteroid medications such as
prednisone and other drugs that suppress the immune system, helps heal the blisters
and alleviates itching.9 However, BP can be life-threatening, especially for elderly individuals in
poor health.10 Only a few cases of KS associated with immunosuppressive therapy for
dermatological diseases have been reported in the literature.11–18 Due to the low incidence of
these diseases, there is a lack of retrospective studies in the Chinese population
on the frequency of KS in the patients with bullous diseases. Interestingly, the
occurrence of classic KS in the Xinjiang Uygur Autonomous Region, the northwest part
of China, is high probably due to the population being composed of 59.43% minority
ethnic groups.5 According to HLA-B gene analysis, some of these people
originated in Europe 3000 years ago, migrated to Central Asia and Xinjiang, where
they then merged with the local ethnic groups and evolved into the current
race.19,20
Kaposi’s sarcoma is caused by HHV-8 infection, which is considered to be a triggering
factor for bullous diseases, especially pemphigus.21,22 However, the association
between HHV-8 and BP remains controversial.23 A recent retrospective study in non-HIV KS patients reported that 14 (1.03%)
of 1362 patients with classic or iatrogenic KS also suffered from bullous disease.24 Half of these 14 patients developed KS after the onset of bullous disease,
whereas the other half developed a bullous disease after being diagnosed with KS.24 Only a few cases on concurrent HHV-8 and BP have been reported in the
literature.12,21,22 HHV-8 identification in KS with BP has been reported only in
one patient in the Chinese population.25
There are discrepancies pertaining to HHV-8 positivity in patients with KS-associated
BP.12,14 In 2001, a
report of a case of KS after immunosuppressive therapy for BP demonstrated that
serum HHV-8 protein was detectable 4 months before the appearance of KS and
initiation of prednisolone therapy.11 The findings in this previous case suggested that the immunosuppression
associated with the treatment for BP possibly activated a latent HHV-8 infection and
induced the development of KS.11 The current patient tested positive for HHV-8 on a biopsy of the lesion. The
experience with this current patient suggests that HHV-8 positivity may be an
indicator for potential development of KS in BP patients. However, it is not clear
whether HHV-8 induced BP and subsequently KS. To clarify the sequence and causation,
it is necessary to compare the incidence of HHV-8-positive and HHV-8-negative KS
appearing after the onset of BP.
Prednisolone and azathioprine are the drugs most frequently implicated in
immunosuppression-associated KS.15 It is difficult to evaluate the relationship between the type, dosage and
duration of immunosuppressive medication and the onset of KS. The mean time to KS
onset, according to the current literature, is 4 months from the initiation
immunosuppressive therapy.26 In this current patient, BP treatment had been administered for 7 months
before the onset of KS.
Declaration of conflicting interest: The authors declare that there are no conflicts of interest.
Funding: This research received no specific grant from any funding agency in the public,
commercial, or not-for-profit sectors.
ORCID iD: Xiao-Qun Luo https://orcid.org/0000-0002-5345-6517 | Oral | DrugAdministrationRoute | CC BY-NC | 33496629 | 19,437,819 | 2021-01 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Acute lymphocytic leukaemia recurrent'. | Early autologous and allogeneic peripheral blood stem cell transplantation for adult patients with acute B and T cell precursor neoplasms: a 12-year single center experience.
Adult acute lymphoblastic leukemia/lymphoma (ALL/LBL) is a rare and heterogeneous malignancy characterized by uncontrolled proliferation of B or T cell precursor cells. Here, we retrospectively analyzed the outcome of early autologous stem cell transplantation in standard-risk patients in first complete remission (n=24) and of allogeneic transplantation in high and highest risk, and relapsed/refractory patients (n=35). The 10-year overall survival after autologous transplantation was 45%. The 10-year overall survival after allogeneic transplantation was 58%. The cumulative incidence of relapse was 29% after allogeneic and 67% after autologous transplantation. The cumulative incidence of non-relapse mortality was 0% after autologous and 12% after allogeneic transplantation. This retrospective single center analysis in a limited number of standard-risk patients clearly demonstrates that early autologous transplantation in first complete remission leads to an acceptable long-term outcome with a short overall treatment duration of less than 6 months compared with more than 2 years with conventional chemotherapy. More sensitive and standardized methods to detect minimal residual disease (MRD) will further help to identify those patients more accurately who are most likely to benefit from such a short and intensive treatment strategy (i.e., MRD negative standard-risk patients) or those who require early targeted therapy (e.g., blinatumomab) in case of MRD positivity. Early allogeneic transplantation results in long-term survival/cure in nearly two-thirds of all high and highest risk, and relapsed/refractory patients.
Introduction
Acute lymphoblastic leukemia/lymphoma (ALL/LBL) is a rare and heterogeneous hematological malignancy. Uncontrolled proliferation of lymphoid progenitor cells committed to the B or T cell lineage in the bone marrow, the peripheral blood, and/or lymphatic and extra-lymphatic tissue characterize the disease. The estimated age- and disease-dependent annual incidence in Europe ranges between 0.17 and 1.45 per 100,000 individuals [1]. Over the past decades, treatment results have improved due to optimized risk stratification, the implementation of monoclonal antibodies (rituximab), BiTE antibodies (blinatumomab), and antibody-drug conjugates (inotuzumab-ozogamicin), as well as the development of highly sensitive diagnostic tools for MRD assessment [2–6]. Besides patient- and disease-related risk-factors and response dynamics, MRD has evolved to the most sensitive marker for the risk of relapse after conventional chemotherapy as well as after stem cell transplantation (SCT) [1, 4, 7]. Allogeneic stem cell transplantation (SCT) is the treatment of choice for high and highest risk as well as relapsed/refractory patients [2, 3]. The conventional treatment approach for standard-risk patients consists of remission induction, consolidation, intensification, and maintenance with an overall treatment duration of 2 or more years [4]. The regimens most commonly used for adults in Europe are based on the pediatric BFM (Berlin-Frankfurt-Münster) protocol, whereas in the USA and many other parts of the world, the hyper-CVAD regimen is preferred [1, 8]. Despite such intensive treatment algorithms, results for adult ALL patients remain unsatisfying [4]. The role of autologous SCT (ASCT) in order to improve treatment results remains controversial. Clinical trials including randomized studies in the pre-MRD era in unselected ALL cohorts have failed to demonstrate any beneficial effect of such a treatment approach [2, 5]. However, studies that are more recent have shown that ASCT might be a suitable option at least for specific subgroups [6, 7].
The main goal of our retrospective single center analysis was to assess whether a significant reduction of treatment duration by early consolidation with ASCT in first complete remission in adult standard-risk ALL is feasible without loss of efficacy compared with the usually recommended conventional post-remission consolidation/maintenance strategy with an overall treatment duration of 2.5–3 years.
Patients and methods
Patients
Between March 2008 and October 2019, fifty-nine adult patients with newly diagnosed B- or T-precursor ALL/LBL received either an autologous (n=24) or allogeneic (n=35) SCT. All patients gave written informed consent. Virtually all patients including those with T-LBL received remission induction (Phase I and II) and consolidation I according to the German Multicenter ALL Protocol for Adults (GMALL 07/2003 Amendment 6, 30.06.2010; Consensus Guidelines of the German Multicenter Study Group for the Treatment of T-lymphoblastic Lymphoma in Adults, Version 1, 18.01.2011) or in patients > 55 years of age according to the GMALL Elderly Protocol 1/2003 (Amendment 5, 21.06.10). Patients with Philadelphia chromosome positive (Ph+) ALL additionally received imatinib 600 mg daily according to the respective protocol.
Characteristics of patients receiving a first ASCT are listed in Table 1. All patients with B- or T-precursor ALL had standard-risk disease as defined by the GMALL 07/2003 protocol. Autologous peripheral blood stem cells were mobilized with granulocyte-colony stimulating factor (G-CSF) 10–20 μg/day starting on day 10 after the first consolidation cycle according to the GMALL 07/2003 protocol. All stem cell products were tested leukemia-free based on conventional fluorescence-activated cell sorter (FACS) analysis, conventional cytogenetics including fluorescence in-situ hybridization (FISH) analysis, and Ig/TCR clonality GeneScan analysis. Median time from diagnosis to transplant in patients in first complete remission (CR1) was 4.9 (range, 2.9–8.9) months. Only one patient received the transplant 20.2 months after diagnosis in second complete remission (CR2) because of a life-threatening infection during consolidation I. Complete remission was determined by bone marrow examination including cytology, conventional FACS analysis, conventional cytogenetics, FISH analysis, real-time polymerase-chain reaction (RT-PCR), Ig/TCR clonality GeneScan analysis, and/or positron emission tomography/computed tomography (PET/CT) scan and/or cerebrospinal fluid assessment when indicated. Additionally, since 2019 next-generation sequencing (NGS) and since April 2019 MRD diagnostics by next-generation flow cytometry (NGF), according to EuroFlow consensus, recommendations (sensitivity < 10−5) were available. According to The GMALL recommendations for stem cell transplantation, all patients received a conditioning regimen with fractionated total body irradiation (fTBI, 12 Gray) combined with either cyclophosphamide (120 mg/m2) or etoposide (60 mg/kg). One patient refusing total body irradiation received intravenous busulfan (12.8 mg/kg) and melphalan (140 mg/m2). All patients with T-LBL (n=7) received the BEAM (carmustine, etoposide, cytarabine, melphalan) regimen [9, 10].Table 1 Characteristics of standard-risk patients receiving an autologous stem cell transplantation in first or second complete remission (n=24)
Median age at the time of diagnosis (years, range) 28 (18–66)
Male/female ratio 16:8
Median time from diagnosis to ASCT in patients in CR1 (months, range) 4.9 (2.9–8.9)
ALL subclassification
B precursor, Ph− 14
T precursor 3
T-LBL 7
Remission status at the time of ASCT
CR1 23
CR2 1
Conditioning regimen
TBI-containing (12 gray) 16
BUMEL 1
BEAM 7
Median CD34 cell number (×106/kg BW, range) 3.74 (1.17–20.87)
Median number of days to leukocyte engraftment (≥ 1.0 G/L) 11 (9–14)
Abbreviations: ALL, acute lymphoblastic leukemia; Ph−, Philadelphia chromosome negative; LBL, lymphoblastic lymphoma; CR1, first complete remission; CR2, second complete remission; TBI, total body irradiation; BUMEL, busulfan, melphalan; BEAM, carmustine, etoposide, cytarabine, melphalan; BW, body weight
Characteristics of patients receiving a first allogeneic SCT are listed in Table 2. All patients receiving an allogeneic SCT in CR1 had high- or highest-risk disease as defined by the GMALL 07/2003 protocol and received remission induction and consolidation according to this protocol (n=29, 83%). One patient was in CR2, and five patients had relapsed/refractory disease. Median time from diagnosis to SCT was 4.5 (range, 1.2–14.8) months. The majority of the patients (83%) had precursor B cell ALL with seven patients being Philadelphia chromosome positive (Ph1+). Conditioning was TBI-containing (fTBI, 12 Gray) and myeloablative in 22/35 (63%) patients, and of fludarabine-based reduced intensity with or without lower doses of fractionated TBI (fTBI, 8 Gray) in the remaining 13/35 (37%) patients. Graft-versus-host disease prophylaxis was calcineurin inhibitor based combined with either methotrexate in the myeloablative setting (n=22) or with mycophenolate mofetil in the reduced-intensity setting (n=13). Eighteen patients (51%) received a graft (mainly peripheral blood stem cells) from an HLA-identical sibling donor. Twelve patients (34%) received allogeneic transplants from HLA-matched unrelated donors (10/10). Five patients (14%) receiving a graft from HLA-mismatched unrelated donors additionally received anti-thymocyte globulin (Grafalon®, 30–60 mg/kg) prior to transplant. All but one patient who died on day+1 because of septic shock engrafted after a median of 13 (range, 8–21) days.Table 2 Characteristics of high- and highest-risk and relapsed/refractory patients receiving an allogeneic stem cell transplantation (n=35)
Median age at the time of diagnosis (years, range) 44 (20–70)
Male/female ratio 13:22
Median time from diagnosis to SCT (months, range) 4.5 (1.2–14.8)
ALL subclassification
B precursor, Ph− 22
B precursor, Ph+ 7
T precursor 6
Remission status at the time of SCT
CR1 29
CR2 1
R/R 5
Conditioning regimen
Myeloablative TBI-containing (12 gray) 22
Reduced-intensity
TBI-containing (8 Gray) 4
Chemotherapy only 9
Stem cell donor
HLA-ID sibling 18
Matched unrelated (10/10) 12
1-Ag mismatched unrelated 5
Median donor age (years, range) 33 (13–66)
Stem cell source
BM 2
PB 33
Recipient/donor sex combination
m/f 7
Others 28
Median CD34 cell number (×106/kg BW, range) 6.15 (2.91–18.9)
Median number of days to leukocyte engraftment (≥ 1.0 G/L) 13 (8–21)
Abbreviations: SCT, stem cell transplantation; ALL, acute lymphoblastic leukemia; Ph−, Philadelphia chromosome negative; Ph+, Philadelphia chromosome positive; CR, complete remission; R/R, relapsed/refractory; TBI, total body irradiation; HLA-ID, human leukocyte antigen-identical; AG, antigen; BM, bone marrow; PB, peripheral blood; m/f, male/female; BW, body weight
Thirteen patients in the autologous group relapsed after a median of 8.2 (range, 2.8–77.1) months, with only two patients having late relapses > 2 years after ASCT. Twelve patients proceeded to an allogeneic SCT within a median of 1.9 (1.3–3.2) months. All but one patient received fludarabine-, clofarabine-, or nelarabine-based salvage therapy prior to transplant [11, 12]. Only four patients achieved a CR2 or CR3. All other patients (8/12, 67%) had refractory disease at the time of allogeneic SCT. Conditioning was of reduced intensity in all patients (fludarabine/busulfan/melphalan (FBM), n=8; thiotepa/busulfan/fludarabine, n=4). All but two patients who died too early due to septic multi-organ failure engrafted after a median of 13 (range 8–23) days.
Study endpoints
The primary endpoint of this retrospective single center analysis was overall survival (OS). Secondary endpoints were disease-free survival (DFS), non-relapse mortality (NRM), and relapse incidence (RI).
Statistical methods
Data were retrospectively reviewed and analyzed as of February 2020. All statistics were computed using NCSS Statistical Software version 19.0.5. The probabilities of OS were calculated using the Kaplan-Meier method from the date of the first transplant until death. DFS was calculated from the date of the first transplant until relapse or death whichever occurred first. The cumulative RI was calculated from the date of the first transplant until relapse with death without relapse as competing risk. The cumulative incidence of NRM was calculated from the date of the first transplant to the date of death without prior relapse with death from relapse as competing risk. NRM for patients receiving an allogeneic SCT because of relapse after a first ASCT was calculated until the date of the allogeneic SCT [13].
Results
Autologous peripheral blood stem cell transplantation in adult standard-risk patients with B- and T-precursor ALL/LBL in first complete remission
The 10-year OS and DFS for the entire cohort was 45% (95% CI, 23–68%) and 33% (95% CI, 8–58%), respectively (Fig. 1a and b). One patient died a natural death more than 10 years after ASCT in ongoing complete remission. The NRM after ASCT was 0% (Fig. 1c). Thirteen patients relapsed with the original leukemic clone with two patients having late relapses > 2 years after ASCT. The resulting cumulative RI at 5 and 8 years after ASCT was 56% (95% CI, 38–83%) and 67% (95% CI, 47–97%), respectively (Fig. 1d).Fig. 1 Autologous stem cell transplantation in adult standard-risk patients with B- and T-precursor ALL/LBL in first complete remission. a) overall survival b) disease-free survival c) cumulative incidence of non-relapse mortality d) cumulative incidence of relapse
Allogeneic stem cell transplantation in high- and highest-risk patients in first complete remission or relapsed/refractory B- and T-precursor ALL/LBL
The 10-year OS for the entire cohort was 58% (95% CI, 40–76%) (Fig. 2a). The 10-year DFS was 55% (95% CI, 37–73%) (Fig. 2b). The 2- and 5-year NRM was 12% (95% CI, 5–30%) (Fig. 2c). This apparently low NRM most likely resulted from the relatively high proportion of reduced-intensity conditioning used for the transplants (37%). Although NRM was higher in the myeloablative setting (15%; 95% CI, 5–42%) compared with reduced-intensity conditioning (8%; 95% CI, 1–50%), this difference was statistically not significant due to low patient number. Nine patients relapsed within the first 2 years after transplant and the cumulative RI was 30% (95% CI, 17–53%) (Fig. 2d). No relapse occurred > 2 years from transplant.Fig. 2 Allogeneic stem cell transplantation in high-and highest-risk patients in first complete remission or relapsed/refractory B- and T-precursor ALL/LBL. a) overall survival b) disease-free survival c) cumulative incidence of non-relapse mortality d) cumulative incidence of relapse
Twelve (34%) patients developed acute graft-versus-host disease ≥ grade II resulting in a significantly better 5- and 10-year OS (92% vs 42%, p = 0.02) and DFS (79% vs 42%, p = 0.05) mainly due to a trend towards a lower RI compared with patients without acute graft-versus-host disease (13% vs 38%, p = 0.07). There was no difference in NRM between patients with or without acute graft-versus-host disease ≥ grade II (data not shown).
Chronic graft-versus-host disease occurred in 7/32 (22%) patients at risk surviving > 100 days after transplant. There were trends towards a better OS (71% vs 55%, p = 0.33) and a better DFS (71% vs 50%, p = 0.26) due to a lower RI (0% vs 39%, p = 0.06), with no difference in NRM (14% vs 11%) for patients with chronic graft-versus-host disease compared with patients without chronic graft-versus-host disease.
Moreover, no significant differences in outcome and RI of patients with T cell precursor or Ph+ ALL compared with other B-precursor subtypes were observed after allogeneic SCT (data not shown).
Allogeneic stem cell transplantation for relapse after ASCT
The outcome of patients receiving an allogeneic SCT for relapse after ASCT (n=12) was dismal with only two patients surviving > 2 years (OS 11%; 95% CI, 0–38%). It is of note that the majority of patients (8/12; 67%) were refractory to either fludarabine-, clofarabine-, or nelarabine-based salvage regimen prior to allogeneic SCT. As expected, this poor outcome was mainly due to a high 1-year RI of 50% (95% CI, 28–88%) and a high 1-year NRM of 33% (95% CI, 15–74%).
Discussion
This retrospective single center analysis demonstrates that early intensification with ASCT in CR1 in adult patients with standard-risk B- or T-precursor ALL/LBL achieves a long-term OS and DFS of 45% and 33%, respectively, without any treatment-related mortality. Although heterogeneous due to the retrospective nature of our analysis, the distribution of the immunologic subtypes corresponds to that observed within the GMALL studies, and all patients including those with T-LBL received the same induction and intensive consolidation according to the GMALL 07/2003 protocol and the GMALL consensus statement for the treatment of adult patients with T-LBL. Despite the short overall treatment duration with a median time from diagnosis to transplant of 4.9 months, our survival data are comparable to the results of the LALA-94 trial in which standard-risk patients receiving remission induction, consolidation, and a consecutive maintenance program for 2 years achieved a 5-year OS and DFS of 44% and 35%, respectively [14]. In the MRC UKALLXII/ECOG E2993, multinational trial patients were randomized by a donor versus no donor stratification after remission induction and 3 courses of high-dose methotrexate to either receive an allogeneic transplant or, if no donor was available, to receive either consolidation/maintenance or an autograft. Among those patients randomized to chemotherapy versus autograft, the 5-year OS in standard-risk patients was 46% with chemotherapy and only 37% with an autograft [5]. Better results were reported by the PETHEMA ALL-AR-03 trial, in which only high-risk patients were eligible. Patients with good early cytological response and MRD negativity at the end of consolidation were allocated to delayed consolidation/maintenance therapy instead of allogeneic SCT. The 5-year OS and DFS of these MRD negative high-risk patients was 58% and 52%, respectively, particularly emphasizing the importance of highly sensitive MRD measurements for treatment guidance [15].
Whether MRD negativity determined by highly sensitive tools can also be used to significantly reduce the overall treatment duration by early intensification with ASCT in standard-risk patients in CR1 remains an open question. Besides the limited patient number, the fact that NGF for highly sensitive MRD detection was not available in our institution before April 2019 is another major drawback of our retrospective single center analysis [16]. Minimal residual disease negativity by NGF was therefore only documented in 1/1 patient in the autogroup and in 4/5 patients in the allogroup after this diagnostic tool has been implemented. However, although no firm conclusion can be drawn for such small patient numbers, we are convinced that the implementation of NGF will further improve the outcome of our early autotransplant strategy especially by reducing the risk of relapse through an optimized selection of those standard-risk patients who are most likely to benefit from such a short and intensive treatment approach, i.e., MRD negative patients. Furthermore, standard-risk patients who do not achieve MRD negativity (MolCR as defined by ESMO Guidelines) are candidates for one or two early courses of blinatumomab [1]. With blinatumomab, MRD negativity can be achieved in another 70–80% of MRD positive patients as has been demonstrated by the BLAST trial [17]. Whether in blinatumomab responders a consolidation with an autologous transplant with low or even no mortality, as reported here, would be preferable over allogeneic transplantation at least in standard-risk patients because of the high NRM of 37% seen in the BLAST Trial remains an open question [18]. Moreover, it remains elusive whether successful stem cell mobilization is feasible after the administration of blinatumomab to enable subsequent autotransplant at all [19].
Relapse of the underlying disease was the sole reason for the failure of our concept of a short and intensive treatment duration with a final autotransplant. The majority of these relapses were refractory to standard of care salvage chemotherapy thereby resulting in a poor outcome even after subsequent allogeneic stem cell transplantation. Our data in this patient cohort is comparable with the recently published report from the EBMT Acute Leukemia Working Party on the results of allogeneic stem cell transplantation with sequential conditioning using the FLAMSA-RIC strategy in adult patients with refractory or relapsed acute lymphoblastic leukemia with an overall survival of only 17% and a graft-versus-host disease- and leukemia-free survival of only 14% at 2 years [20]. Better results are only obtainable when MRD negativity prior to allotransplant is achieved by implementing targeted therapies [21, 22]. Whether a subsequent allogeneic SCT can be omitted in MRD responders after blinatumomab in the relapsed/refractory setting as it has been supposed by the results of the phase 3 TOWER study has to be interpreted with caution [22].
While the use of ASCT in adult ALL remains a matter of debate, frontline allogeneic SCT is the treatment of choice for Ph1+ and other high-risk B- and T-precursor neoplasms [23–25]. Our results in thirty-five high- and highest-risk and relapsed/refractory patients perfectly fit to the recently published retrospective data of the Acute Leukemia Working Party of the European Society for Blood and Marrow Transplantation (EBMT) showing a long-term survival/cure in about two-thirds of the patients, a relatively low NRM of < 20%, and a RI of about 30% [26]. Whether in the era of potent tyrosine kinase inhibitors (TKIs) in combination with BiTE antibodies or antibody-drug conjugates allogeneic SCT can be replaced by ASCT with less morbidity and mortality at least in those high- and highest-risk patients achieving a deep MolCR remains to be shown by prospective randomized trials.
In conclusion, despite all limitations, our retrospective single center analysis clearly demonstrates that an early autotransplant strategy significantly shortens treatment duration while providing an acceptable long-term outcome in standard-risk patients in first complete remission. The implementation of NGF might be helpful to better define those patients who would benefit most from such a short and intensive treatment approach by reducing the risk of relapse and the consecutive need for a salvage allotransplant with dismal outcome. The increasing availability of MRD evaluation by standardized, highly sensitive diagnostic tools and potent-targeted therapies (e.g., blinatumomab, TKIs) not only allows a more individualized medicine in adult ALL but also asks for the need to reconsider and redefine the role of prolonged chemotherapy versus early autologous or allogeneic stem cell transplantation in the treatment algorithms of various ALL subgroups.
The authors would like to acknowledge all SCT recipients and donors as well as the nursing team for the excellent state of art clinical care given to the patients.
Funding
Open Access funding provided by University of Innsbruck and Medical University of Innsbruck.
Data availability
The datasets generated or analyzed during this study are available from the corresponding author on reasonable request.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflicts of interest.
Ethics approval
Ethical approval was waived by the local Ethics Committee of Innsbruck Medical University of Innsbruck in view of the retrospective nature of the study and all the procedures being performed were part of the routine care.
Consent to participate
Informed consent was obtained from all patients included in the study.
Consent for publication
Patients signed informed consent regarding publishing their data.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | FILGRASTIM | DrugsGivenReaction | CC BY | 33496839 | 18,877,749 | 2021-03 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Death'. | Early autologous and allogeneic peripheral blood stem cell transplantation for adult patients with acute B and T cell precursor neoplasms: a 12-year single center experience.
Adult acute lymphoblastic leukemia/lymphoma (ALL/LBL) is a rare and heterogeneous malignancy characterized by uncontrolled proliferation of B or T cell precursor cells. Here, we retrospectively analyzed the outcome of early autologous stem cell transplantation in standard-risk patients in first complete remission (n=24) and of allogeneic transplantation in high and highest risk, and relapsed/refractory patients (n=35). The 10-year overall survival after autologous transplantation was 45%. The 10-year overall survival after allogeneic transplantation was 58%. The cumulative incidence of relapse was 29% after allogeneic and 67% after autologous transplantation. The cumulative incidence of non-relapse mortality was 0% after autologous and 12% after allogeneic transplantation. This retrospective single center analysis in a limited number of standard-risk patients clearly demonstrates that early autologous transplantation in first complete remission leads to an acceptable long-term outcome with a short overall treatment duration of less than 6 months compared with more than 2 years with conventional chemotherapy. More sensitive and standardized methods to detect minimal residual disease (MRD) will further help to identify those patients more accurately who are most likely to benefit from such a short and intensive treatment strategy (i.e., MRD negative standard-risk patients) or those who require early targeted therapy (e.g., blinatumomab) in case of MRD positivity. Early allogeneic transplantation results in long-term survival/cure in nearly two-thirds of all high and highest risk, and relapsed/refractory patients.
Introduction
Acute lymphoblastic leukemia/lymphoma (ALL/LBL) is a rare and heterogeneous hematological malignancy. Uncontrolled proliferation of lymphoid progenitor cells committed to the B or T cell lineage in the bone marrow, the peripheral blood, and/or lymphatic and extra-lymphatic tissue characterize the disease. The estimated age- and disease-dependent annual incidence in Europe ranges between 0.17 and 1.45 per 100,000 individuals [1]. Over the past decades, treatment results have improved due to optimized risk stratification, the implementation of monoclonal antibodies (rituximab), BiTE antibodies (blinatumomab), and antibody-drug conjugates (inotuzumab-ozogamicin), as well as the development of highly sensitive diagnostic tools for MRD assessment [2–6]. Besides patient- and disease-related risk-factors and response dynamics, MRD has evolved to the most sensitive marker for the risk of relapse after conventional chemotherapy as well as after stem cell transplantation (SCT) [1, 4, 7]. Allogeneic stem cell transplantation (SCT) is the treatment of choice for high and highest risk as well as relapsed/refractory patients [2, 3]. The conventional treatment approach for standard-risk patients consists of remission induction, consolidation, intensification, and maintenance with an overall treatment duration of 2 or more years [4]. The regimens most commonly used for adults in Europe are based on the pediatric BFM (Berlin-Frankfurt-Münster) protocol, whereas in the USA and many other parts of the world, the hyper-CVAD regimen is preferred [1, 8]. Despite such intensive treatment algorithms, results for adult ALL patients remain unsatisfying [4]. The role of autologous SCT (ASCT) in order to improve treatment results remains controversial. Clinical trials including randomized studies in the pre-MRD era in unselected ALL cohorts have failed to demonstrate any beneficial effect of such a treatment approach [2, 5]. However, studies that are more recent have shown that ASCT might be a suitable option at least for specific subgroups [6, 7].
The main goal of our retrospective single center analysis was to assess whether a significant reduction of treatment duration by early consolidation with ASCT in first complete remission in adult standard-risk ALL is feasible without loss of efficacy compared with the usually recommended conventional post-remission consolidation/maintenance strategy with an overall treatment duration of 2.5–3 years.
Patients and methods
Patients
Between March 2008 and October 2019, fifty-nine adult patients with newly diagnosed B- or T-precursor ALL/LBL received either an autologous (n=24) or allogeneic (n=35) SCT. All patients gave written informed consent. Virtually all patients including those with T-LBL received remission induction (Phase I and II) and consolidation I according to the German Multicenter ALL Protocol for Adults (GMALL 07/2003 Amendment 6, 30.06.2010; Consensus Guidelines of the German Multicenter Study Group for the Treatment of T-lymphoblastic Lymphoma in Adults, Version 1, 18.01.2011) or in patients > 55 years of age according to the GMALL Elderly Protocol 1/2003 (Amendment 5, 21.06.10). Patients with Philadelphia chromosome positive (Ph+) ALL additionally received imatinib 600 mg daily according to the respective protocol.
Characteristics of patients receiving a first ASCT are listed in Table 1. All patients with B- or T-precursor ALL had standard-risk disease as defined by the GMALL 07/2003 protocol. Autologous peripheral blood stem cells were mobilized with granulocyte-colony stimulating factor (G-CSF) 10–20 μg/day starting on day 10 after the first consolidation cycle according to the GMALL 07/2003 protocol. All stem cell products were tested leukemia-free based on conventional fluorescence-activated cell sorter (FACS) analysis, conventional cytogenetics including fluorescence in-situ hybridization (FISH) analysis, and Ig/TCR clonality GeneScan analysis. Median time from diagnosis to transplant in patients in first complete remission (CR1) was 4.9 (range, 2.9–8.9) months. Only one patient received the transplant 20.2 months after diagnosis in second complete remission (CR2) because of a life-threatening infection during consolidation I. Complete remission was determined by bone marrow examination including cytology, conventional FACS analysis, conventional cytogenetics, FISH analysis, real-time polymerase-chain reaction (RT-PCR), Ig/TCR clonality GeneScan analysis, and/or positron emission tomography/computed tomography (PET/CT) scan and/or cerebrospinal fluid assessment when indicated. Additionally, since 2019 next-generation sequencing (NGS) and since April 2019 MRD diagnostics by next-generation flow cytometry (NGF), according to EuroFlow consensus, recommendations (sensitivity < 10−5) were available. According to The GMALL recommendations for stem cell transplantation, all patients received a conditioning regimen with fractionated total body irradiation (fTBI, 12 Gray) combined with either cyclophosphamide (120 mg/m2) or etoposide (60 mg/kg). One patient refusing total body irradiation received intravenous busulfan (12.8 mg/kg) and melphalan (140 mg/m2). All patients with T-LBL (n=7) received the BEAM (carmustine, etoposide, cytarabine, melphalan) regimen [9, 10].Table 1 Characteristics of standard-risk patients receiving an autologous stem cell transplantation in first or second complete remission (n=24)
Median age at the time of diagnosis (years, range) 28 (18–66)
Male/female ratio 16:8
Median time from diagnosis to ASCT in patients in CR1 (months, range) 4.9 (2.9–8.9)
ALL subclassification
B precursor, Ph− 14
T precursor 3
T-LBL 7
Remission status at the time of ASCT
CR1 23
CR2 1
Conditioning regimen
TBI-containing (12 gray) 16
BUMEL 1
BEAM 7
Median CD34 cell number (×106/kg BW, range) 3.74 (1.17–20.87)
Median number of days to leukocyte engraftment (≥ 1.0 G/L) 11 (9–14)
Abbreviations: ALL, acute lymphoblastic leukemia; Ph−, Philadelphia chromosome negative; LBL, lymphoblastic lymphoma; CR1, first complete remission; CR2, second complete remission; TBI, total body irradiation; BUMEL, busulfan, melphalan; BEAM, carmustine, etoposide, cytarabine, melphalan; BW, body weight
Characteristics of patients receiving a first allogeneic SCT are listed in Table 2. All patients receiving an allogeneic SCT in CR1 had high- or highest-risk disease as defined by the GMALL 07/2003 protocol and received remission induction and consolidation according to this protocol (n=29, 83%). One patient was in CR2, and five patients had relapsed/refractory disease. Median time from diagnosis to SCT was 4.5 (range, 1.2–14.8) months. The majority of the patients (83%) had precursor B cell ALL with seven patients being Philadelphia chromosome positive (Ph1+). Conditioning was TBI-containing (fTBI, 12 Gray) and myeloablative in 22/35 (63%) patients, and of fludarabine-based reduced intensity with or without lower doses of fractionated TBI (fTBI, 8 Gray) in the remaining 13/35 (37%) patients. Graft-versus-host disease prophylaxis was calcineurin inhibitor based combined with either methotrexate in the myeloablative setting (n=22) or with mycophenolate mofetil in the reduced-intensity setting (n=13). Eighteen patients (51%) received a graft (mainly peripheral blood stem cells) from an HLA-identical sibling donor. Twelve patients (34%) received allogeneic transplants from HLA-matched unrelated donors (10/10). Five patients (14%) receiving a graft from HLA-mismatched unrelated donors additionally received anti-thymocyte globulin (Grafalon®, 30–60 mg/kg) prior to transplant. All but one patient who died on day+1 because of septic shock engrafted after a median of 13 (range, 8–21) days.Table 2 Characteristics of high- and highest-risk and relapsed/refractory patients receiving an allogeneic stem cell transplantation (n=35)
Median age at the time of diagnosis (years, range) 44 (20–70)
Male/female ratio 13:22
Median time from diagnosis to SCT (months, range) 4.5 (1.2–14.8)
ALL subclassification
B precursor, Ph− 22
B precursor, Ph+ 7
T precursor 6
Remission status at the time of SCT
CR1 29
CR2 1
R/R 5
Conditioning regimen
Myeloablative TBI-containing (12 gray) 22
Reduced-intensity
TBI-containing (8 Gray) 4
Chemotherapy only 9
Stem cell donor
HLA-ID sibling 18
Matched unrelated (10/10) 12
1-Ag mismatched unrelated 5
Median donor age (years, range) 33 (13–66)
Stem cell source
BM 2
PB 33
Recipient/donor sex combination
m/f 7
Others 28
Median CD34 cell number (×106/kg BW, range) 6.15 (2.91–18.9)
Median number of days to leukocyte engraftment (≥ 1.0 G/L) 13 (8–21)
Abbreviations: SCT, stem cell transplantation; ALL, acute lymphoblastic leukemia; Ph−, Philadelphia chromosome negative; Ph+, Philadelphia chromosome positive; CR, complete remission; R/R, relapsed/refractory; TBI, total body irradiation; HLA-ID, human leukocyte antigen-identical; AG, antigen; BM, bone marrow; PB, peripheral blood; m/f, male/female; BW, body weight
Thirteen patients in the autologous group relapsed after a median of 8.2 (range, 2.8–77.1) months, with only two patients having late relapses > 2 years after ASCT. Twelve patients proceeded to an allogeneic SCT within a median of 1.9 (1.3–3.2) months. All but one patient received fludarabine-, clofarabine-, or nelarabine-based salvage therapy prior to transplant [11, 12]. Only four patients achieved a CR2 or CR3. All other patients (8/12, 67%) had refractory disease at the time of allogeneic SCT. Conditioning was of reduced intensity in all patients (fludarabine/busulfan/melphalan (FBM), n=8; thiotepa/busulfan/fludarabine, n=4). All but two patients who died too early due to septic multi-organ failure engrafted after a median of 13 (range 8–23) days.
Study endpoints
The primary endpoint of this retrospective single center analysis was overall survival (OS). Secondary endpoints were disease-free survival (DFS), non-relapse mortality (NRM), and relapse incidence (RI).
Statistical methods
Data were retrospectively reviewed and analyzed as of February 2020. All statistics were computed using NCSS Statistical Software version 19.0.5. The probabilities of OS were calculated using the Kaplan-Meier method from the date of the first transplant until death. DFS was calculated from the date of the first transplant until relapse or death whichever occurred first. The cumulative RI was calculated from the date of the first transplant until relapse with death without relapse as competing risk. The cumulative incidence of NRM was calculated from the date of the first transplant to the date of death without prior relapse with death from relapse as competing risk. NRM for patients receiving an allogeneic SCT because of relapse after a first ASCT was calculated until the date of the allogeneic SCT [13].
Results
Autologous peripheral blood stem cell transplantation in adult standard-risk patients with B- and T-precursor ALL/LBL in first complete remission
The 10-year OS and DFS for the entire cohort was 45% (95% CI, 23–68%) and 33% (95% CI, 8–58%), respectively (Fig. 1a and b). One patient died a natural death more than 10 years after ASCT in ongoing complete remission. The NRM after ASCT was 0% (Fig. 1c). Thirteen patients relapsed with the original leukemic clone with two patients having late relapses > 2 years after ASCT. The resulting cumulative RI at 5 and 8 years after ASCT was 56% (95% CI, 38–83%) and 67% (95% CI, 47–97%), respectively (Fig. 1d).Fig. 1 Autologous stem cell transplantation in adult standard-risk patients with B- and T-precursor ALL/LBL in first complete remission. a) overall survival b) disease-free survival c) cumulative incidence of non-relapse mortality d) cumulative incidence of relapse
Allogeneic stem cell transplantation in high- and highest-risk patients in first complete remission or relapsed/refractory B- and T-precursor ALL/LBL
The 10-year OS for the entire cohort was 58% (95% CI, 40–76%) (Fig. 2a). The 10-year DFS was 55% (95% CI, 37–73%) (Fig. 2b). The 2- and 5-year NRM was 12% (95% CI, 5–30%) (Fig. 2c). This apparently low NRM most likely resulted from the relatively high proportion of reduced-intensity conditioning used for the transplants (37%). Although NRM was higher in the myeloablative setting (15%; 95% CI, 5–42%) compared with reduced-intensity conditioning (8%; 95% CI, 1–50%), this difference was statistically not significant due to low patient number. Nine patients relapsed within the first 2 years after transplant and the cumulative RI was 30% (95% CI, 17–53%) (Fig. 2d). No relapse occurred > 2 years from transplant.Fig. 2 Allogeneic stem cell transplantation in high-and highest-risk patients in first complete remission or relapsed/refractory B- and T-precursor ALL/LBL. a) overall survival b) disease-free survival c) cumulative incidence of non-relapse mortality d) cumulative incidence of relapse
Twelve (34%) patients developed acute graft-versus-host disease ≥ grade II resulting in a significantly better 5- and 10-year OS (92% vs 42%, p = 0.02) and DFS (79% vs 42%, p = 0.05) mainly due to a trend towards a lower RI compared with patients without acute graft-versus-host disease (13% vs 38%, p = 0.07). There was no difference in NRM between patients with or without acute graft-versus-host disease ≥ grade II (data not shown).
Chronic graft-versus-host disease occurred in 7/32 (22%) patients at risk surviving > 100 days after transplant. There were trends towards a better OS (71% vs 55%, p = 0.33) and a better DFS (71% vs 50%, p = 0.26) due to a lower RI (0% vs 39%, p = 0.06), with no difference in NRM (14% vs 11%) for patients with chronic graft-versus-host disease compared with patients without chronic graft-versus-host disease.
Moreover, no significant differences in outcome and RI of patients with T cell precursor or Ph+ ALL compared with other B-precursor subtypes were observed after allogeneic SCT (data not shown).
Allogeneic stem cell transplantation for relapse after ASCT
The outcome of patients receiving an allogeneic SCT for relapse after ASCT (n=12) was dismal with only two patients surviving > 2 years (OS 11%; 95% CI, 0–38%). It is of note that the majority of patients (8/12; 67%) were refractory to either fludarabine-, clofarabine-, or nelarabine-based salvage regimen prior to allogeneic SCT. As expected, this poor outcome was mainly due to a high 1-year RI of 50% (95% CI, 28–88%) and a high 1-year NRM of 33% (95% CI, 15–74%).
Discussion
This retrospective single center analysis demonstrates that early intensification with ASCT in CR1 in adult patients with standard-risk B- or T-precursor ALL/LBL achieves a long-term OS and DFS of 45% and 33%, respectively, without any treatment-related mortality. Although heterogeneous due to the retrospective nature of our analysis, the distribution of the immunologic subtypes corresponds to that observed within the GMALL studies, and all patients including those with T-LBL received the same induction and intensive consolidation according to the GMALL 07/2003 protocol and the GMALL consensus statement for the treatment of adult patients with T-LBL. Despite the short overall treatment duration with a median time from diagnosis to transplant of 4.9 months, our survival data are comparable to the results of the LALA-94 trial in which standard-risk patients receiving remission induction, consolidation, and a consecutive maintenance program for 2 years achieved a 5-year OS and DFS of 44% and 35%, respectively [14]. In the MRC UKALLXII/ECOG E2993, multinational trial patients were randomized by a donor versus no donor stratification after remission induction and 3 courses of high-dose methotrexate to either receive an allogeneic transplant or, if no donor was available, to receive either consolidation/maintenance or an autograft. Among those patients randomized to chemotherapy versus autograft, the 5-year OS in standard-risk patients was 46% with chemotherapy and only 37% with an autograft [5]. Better results were reported by the PETHEMA ALL-AR-03 trial, in which only high-risk patients were eligible. Patients with good early cytological response and MRD negativity at the end of consolidation were allocated to delayed consolidation/maintenance therapy instead of allogeneic SCT. The 5-year OS and DFS of these MRD negative high-risk patients was 58% and 52%, respectively, particularly emphasizing the importance of highly sensitive MRD measurements for treatment guidance [15].
Whether MRD negativity determined by highly sensitive tools can also be used to significantly reduce the overall treatment duration by early intensification with ASCT in standard-risk patients in CR1 remains an open question. Besides the limited patient number, the fact that NGF for highly sensitive MRD detection was not available in our institution before April 2019 is another major drawback of our retrospective single center analysis [16]. Minimal residual disease negativity by NGF was therefore only documented in 1/1 patient in the autogroup and in 4/5 patients in the allogroup after this diagnostic tool has been implemented. However, although no firm conclusion can be drawn for such small patient numbers, we are convinced that the implementation of NGF will further improve the outcome of our early autotransplant strategy especially by reducing the risk of relapse through an optimized selection of those standard-risk patients who are most likely to benefit from such a short and intensive treatment approach, i.e., MRD negative patients. Furthermore, standard-risk patients who do not achieve MRD negativity (MolCR as defined by ESMO Guidelines) are candidates for one or two early courses of blinatumomab [1]. With blinatumomab, MRD negativity can be achieved in another 70–80% of MRD positive patients as has been demonstrated by the BLAST trial [17]. Whether in blinatumomab responders a consolidation with an autologous transplant with low or even no mortality, as reported here, would be preferable over allogeneic transplantation at least in standard-risk patients because of the high NRM of 37% seen in the BLAST Trial remains an open question [18]. Moreover, it remains elusive whether successful stem cell mobilization is feasible after the administration of blinatumomab to enable subsequent autotransplant at all [19].
Relapse of the underlying disease was the sole reason for the failure of our concept of a short and intensive treatment duration with a final autotransplant. The majority of these relapses were refractory to standard of care salvage chemotherapy thereby resulting in a poor outcome even after subsequent allogeneic stem cell transplantation. Our data in this patient cohort is comparable with the recently published report from the EBMT Acute Leukemia Working Party on the results of allogeneic stem cell transplantation with sequential conditioning using the FLAMSA-RIC strategy in adult patients with refractory or relapsed acute lymphoblastic leukemia with an overall survival of only 17% and a graft-versus-host disease- and leukemia-free survival of only 14% at 2 years [20]. Better results are only obtainable when MRD negativity prior to allotransplant is achieved by implementing targeted therapies [21, 22]. Whether a subsequent allogeneic SCT can be omitted in MRD responders after blinatumomab in the relapsed/refractory setting as it has been supposed by the results of the phase 3 TOWER study has to be interpreted with caution [22].
While the use of ASCT in adult ALL remains a matter of debate, frontline allogeneic SCT is the treatment of choice for Ph1+ and other high-risk B- and T-precursor neoplasms [23–25]. Our results in thirty-five high- and highest-risk and relapsed/refractory patients perfectly fit to the recently published retrospective data of the Acute Leukemia Working Party of the European Society for Blood and Marrow Transplantation (EBMT) showing a long-term survival/cure in about two-thirds of the patients, a relatively low NRM of < 20%, and a RI of about 30% [26]. Whether in the era of potent tyrosine kinase inhibitors (TKIs) in combination with BiTE antibodies or antibody-drug conjugates allogeneic SCT can be replaced by ASCT with less morbidity and mortality at least in those high- and highest-risk patients achieving a deep MolCR remains to be shown by prospective randomized trials.
In conclusion, despite all limitations, our retrospective single center analysis clearly demonstrates that an early autotransplant strategy significantly shortens treatment duration while providing an acceptable long-term outcome in standard-risk patients in first complete remission. The implementation of NGF might be helpful to better define those patients who would benefit most from such a short and intensive treatment approach by reducing the risk of relapse and the consecutive need for a salvage allotransplant with dismal outcome. The increasing availability of MRD evaluation by standardized, highly sensitive diagnostic tools and potent-targeted therapies (e.g., blinatumomab, TKIs) not only allows a more individualized medicine in adult ALL but also asks for the need to reconsider and redefine the role of prolonged chemotherapy versus early autologous or allogeneic stem cell transplantation in the treatment algorithms of various ALL subgroups.
The authors would like to acknowledge all SCT recipients and donors as well as the nursing team for the excellent state of art clinical care given to the patients.
Funding
Open Access funding provided by University of Innsbruck and Medical University of Innsbruck.
Data availability
The datasets generated or analyzed during this study are available from the corresponding author on reasonable request.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflicts of interest.
Ethics approval
Ethical approval was waived by the local Ethics Committee of Innsbruck Medical University of Innsbruck in view of the retrospective nature of the study and all the procedures being performed were part of the routine care.
Consent to participate
Informed consent was obtained from all patients included in the study.
Consent for publication
Patients signed informed consent regarding publishing their data.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | FILGRASTIM | DrugsGivenReaction | CC BY | 33496839 | 18,877,749 | 2021-03 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Non-Hodgkin^s lymphoma recurrent'. | Early autologous and allogeneic peripheral blood stem cell transplantation for adult patients with acute B and T cell precursor neoplasms: a 12-year single center experience.
Adult acute lymphoblastic leukemia/lymphoma (ALL/LBL) is a rare and heterogeneous malignancy characterized by uncontrolled proliferation of B or T cell precursor cells. Here, we retrospectively analyzed the outcome of early autologous stem cell transplantation in standard-risk patients in first complete remission (n=24) and of allogeneic transplantation in high and highest risk, and relapsed/refractory patients (n=35). The 10-year overall survival after autologous transplantation was 45%. The 10-year overall survival after allogeneic transplantation was 58%. The cumulative incidence of relapse was 29% after allogeneic and 67% after autologous transplantation. The cumulative incidence of non-relapse mortality was 0% after autologous and 12% after allogeneic transplantation. This retrospective single center analysis in a limited number of standard-risk patients clearly demonstrates that early autologous transplantation in first complete remission leads to an acceptable long-term outcome with a short overall treatment duration of less than 6 months compared with more than 2 years with conventional chemotherapy. More sensitive and standardized methods to detect minimal residual disease (MRD) will further help to identify those patients more accurately who are most likely to benefit from such a short and intensive treatment strategy (i.e., MRD negative standard-risk patients) or those who require early targeted therapy (e.g., blinatumomab) in case of MRD positivity. Early allogeneic transplantation results in long-term survival/cure in nearly two-thirds of all high and highest risk, and relapsed/refractory patients.
Introduction
Acute lymphoblastic leukemia/lymphoma (ALL/LBL) is a rare and heterogeneous hematological malignancy. Uncontrolled proliferation of lymphoid progenitor cells committed to the B or T cell lineage in the bone marrow, the peripheral blood, and/or lymphatic and extra-lymphatic tissue characterize the disease. The estimated age- and disease-dependent annual incidence in Europe ranges between 0.17 and 1.45 per 100,000 individuals [1]. Over the past decades, treatment results have improved due to optimized risk stratification, the implementation of monoclonal antibodies (rituximab), BiTE antibodies (blinatumomab), and antibody-drug conjugates (inotuzumab-ozogamicin), as well as the development of highly sensitive diagnostic tools for MRD assessment [2–6]. Besides patient- and disease-related risk-factors and response dynamics, MRD has evolved to the most sensitive marker for the risk of relapse after conventional chemotherapy as well as after stem cell transplantation (SCT) [1, 4, 7]. Allogeneic stem cell transplantation (SCT) is the treatment of choice for high and highest risk as well as relapsed/refractory patients [2, 3]. The conventional treatment approach for standard-risk patients consists of remission induction, consolidation, intensification, and maintenance with an overall treatment duration of 2 or more years [4]. The regimens most commonly used for adults in Europe are based on the pediatric BFM (Berlin-Frankfurt-Münster) protocol, whereas in the USA and many other parts of the world, the hyper-CVAD regimen is preferred [1, 8]. Despite such intensive treatment algorithms, results for adult ALL patients remain unsatisfying [4]. The role of autologous SCT (ASCT) in order to improve treatment results remains controversial. Clinical trials including randomized studies in the pre-MRD era in unselected ALL cohorts have failed to demonstrate any beneficial effect of such a treatment approach [2, 5]. However, studies that are more recent have shown that ASCT might be a suitable option at least for specific subgroups [6, 7].
The main goal of our retrospective single center analysis was to assess whether a significant reduction of treatment duration by early consolidation with ASCT in first complete remission in adult standard-risk ALL is feasible without loss of efficacy compared with the usually recommended conventional post-remission consolidation/maintenance strategy with an overall treatment duration of 2.5–3 years.
Patients and methods
Patients
Between March 2008 and October 2019, fifty-nine adult patients with newly diagnosed B- or T-precursor ALL/LBL received either an autologous (n=24) or allogeneic (n=35) SCT. All patients gave written informed consent. Virtually all patients including those with T-LBL received remission induction (Phase I and II) and consolidation I according to the German Multicenter ALL Protocol for Adults (GMALL 07/2003 Amendment 6, 30.06.2010; Consensus Guidelines of the German Multicenter Study Group for the Treatment of T-lymphoblastic Lymphoma in Adults, Version 1, 18.01.2011) or in patients > 55 years of age according to the GMALL Elderly Protocol 1/2003 (Amendment 5, 21.06.10). Patients with Philadelphia chromosome positive (Ph+) ALL additionally received imatinib 600 mg daily according to the respective protocol.
Characteristics of patients receiving a first ASCT are listed in Table 1. All patients with B- or T-precursor ALL had standard-risk disease as defined by the GMALL 07/2003 protocol. Autologous peripheral blood stem cells were mobilized with granulocyte-colony stimulating factor (G-CSF) 10–20 μg/day starting on day 10 after the first consolidation cycle according to the GMALL 07/2003 protocol. All stem cell products were tested leukemia-free based on conventional fluorescence-activated cell sorter (FACS) analysis, conventional cytogenetics including fluorescence in-situ hybridization (FISH) analysis, and Ig/TCR clonality GeneScan analysis. Median time from diagnosis to transplant in patients in first complete remission (CR1) was 4.9 (range, 2.9–8.9) months. Only one patient received the transplant 20.2 months after diagnosis in second complete remission (CR2) because of a life-threatening infection during consolidation I. Complete remission was determined by bone marrow examination including cytology, conventional FACS analysis, conventional cytogenetics, FISH analysis, real-time polymerase-chain reaction (RT-PCR), Ig/TCR clonality GeneScan analysis, and/or positron emission tomography/computed tomography (PET/CT) scan and/or cerebrospinal fluid assessment when indicated. Additionally, since 2019 next-generation sequencing (NGS) and since April 2019 MRD diagnostics by next-generation flow cytometry (NGF), according to EuroFlow consensus, recommendations (sensitivity < 10−5) were available. According to The GMALL recommendations for stem cell transplantation, all patients received a conditioning regimen with fractionated total body irradiation (fTBI, 12 Gray) combined with either cyclophosphamide (120 mg/m2) or etoposide (60 mg/kg). One patient refusing total body irradiation received intravenous busulfan (12.8 mg/kg) and melphalan (140 mg/m2). All patients with T-LBL (n=7) received the BEAM (carmustine, etoposide, cytarabine, melphalan) regimen [9, 10].Table 1 Characteristics of standard-risk patients receiving an autologous stem cell transplantation in first or second complete remission (n=24)
Median age at the time of diagnosis (years, range) 28 (18–66)
Male/female ratio 16:8
Median time from diagnosis to ASCT in patients in CR1 (months, range) 4.9 (2.9–8.9)
ALL subclassification
B precursor, Ph− 14
T precursor 3
T-LBL 7
Remission status at the time of ASCT
CR1 23
CR2 1
Conditioning regimen
TBI-containing (12 gray) 16
BUMEL 1
BEAM 7
Median CD34 cell number (×106/kg BW, range) 3.74 (1.17–20.87)
Median number of days to leukocyte engraftment (≥ 1.0 G/L) 11 (9–14)
Abbreviations: ALL, acute lymphoblastic leukemia; Ph−, Philadelphia chromosome negative; LBL, lymphoblastic lymphoma; CR1, first complete remission; CR2, second complete remission; TBI, total body irradiation; BUMEL, busulfan, melphalan; BEAM, carmustine, etoposide, cytarabine, melphalan; BW, body weight
Characteristics of patients receiving a first allogeneic SCT are listed in Table 2. All patients receiving an allogeneic SCT in CR1 had high- or highest-risk disease as defined by the GMALL 07/2003 protocol and received remission induction and consolidation according to this protocol (n=29, 83%). One patient was in CR2, and five patients had relapsed/refractory disease. Median time from diagnosis to SCT was 4.5 (range, 1.2–14.8) months. The majority of the patients (83%) had precursor B cell ALL with seven patients being Philadelphia chromosome positive (Ph1+). Conditioning was TBI-containing (fTBI, 12 Gray) and myeloablative in 22/35 (63%) patients, and of fludarabine-based reduced intensity with or without lower doses of fractionated TBI (fTBI, 8 Gray) in the remaining 13/35 (37%) patients. Graft-versus-host disease prophylaxis was calcineurin inhibitor based combined with either methotrexate in the myeloablative setting (n=22) or with mycophenolate mofetil in the reduced-intensity setting (n=13). Eighteen patients (51%) received a graft (mainly peripheral blood stem cells) from an HLA-identical sibling donor. Twelve patients (34%) received allogeneic transplants from HLA-matched unrelated donors (10/10). Five patients (14%) receiving a graft from HLA-mismatched unrelated donors additionally received anti-thymocyte globulin (Grafalon®, 30–60 mg/kg) prior to transplant. All but one patient who died on day+1 because of septic shock engrafted after a median of 13 (range, 8–21) days.Table 2 Characteristics of high- and highest-risk and relapsed/refractory patients receiving an allogeneic stem cell transplantation (n=35)
Median age at the time of diagnosis (years, range) 44 (20–70)
Male/female ratio 13:22
Median time from diagnosis to SCT (months, range) 4.5 (1.2–14.8)
ALL subclassification
B precursor, Ph− 22
B precursor, Ph+ 7
T precursor 6
Remission status at the time of SCT
CR1 29
CR2 1
R/R 5
Conditioning regimen
Myeloablative TBI-containing (12 gray) 22
Reduced-intensity
TBI-containing (8 Gray) 4
Chemotherapy only 9
Stem cell donor
HLA-ID sibling 18
Matched unrelated (10/10) 12
1-Ag mismatched unrelated 5
Median donor age (years, range) 33 (13–66)
Stem cell source
BM 2
PB 33
Recipient/donor sex combination
m/f 7
Others 28
Median CD34 cell number (×106/kg BW, range) 6.15 (2.91–18.9)
Median number of days to leukocyte engraftment (≥ 1.0 G/L) 13 (8–21)
Abbreviations: SCT, stem cell transplantation; ALL, acute lymphoblastic leukemia; Ph−, Philadelphia chromosome negative; Ph+, Philadelphia chromosome positive; CR, complete remission; R/R, relapsed/refractory; TBI, total body irradiation; HLA-ID, human leukocyte antigen-identical; AG, antigen; BM, bone marrow; PB, peripheral blood; m/f, male/female; BW, body weight
Thirteen patients in the autologous group relapsed after a median of 8.2 (range, 2.8–77.1) months, with only two patients having late relapses > 2 years after ASCT. Twelve patients proceeded to an allogeneic SCT within a median of 1.9 (1.3–3.2) months. All but one patient received fludarabine-, clofarabine-, or nelarabine-based salvage therapy prior to transplant [11, 12]. Only four patients achieved a CR2 or CR3. All other patients (8/12, 67%) had refractory disease at the time of allogeneic SCT. Conditioning was of reduced intensity in all patients (fludarabine/busulfan/melphalan (FBM), n=8; thiotepa/busulfan/fludarabine, n=4). All but two patients who died too early due to septic multi-organ failure engrafted after a median of 13 (range 8–23) days.
Study endpoints
The primary endpoint of this retrospective single center analysis was overall survival (OS). Secondary endpoints were disease-free survival (DFS), non-relapse mortality (NRM), and relapse incidence (RI).
Statistical methods
Data were retrospectively reviewed and analyzed as of February 2020. All statistics were computed using NCSS Statistical Software version 19.0.5. The probabilities of OS were calculated using the Kaplan-Meier method from the date of the first transplant until death. DFS was calculated from the date of the first transplant until relapse or death whichever occurred first. The cumulative RI was calculated from the date of the first transplant until relapse with death without relapse as competing risk. The cumulative incidence of NRM was calculated from the date of the first transplant to the date of death without prior relapse with death from relapse as competing risk. NRM for patients receiving an allogeneic SCT because of relapse after a first ASCT was calculated until the date of the allogeneic SCT [13].
Results
Autologous peripheral blood stem cell transplantation in adult standard-risk patients with B- and T-precursor ALL/LBL in first complete remission
The 10-year OS and DFS for the entire cohort was 45% (95% CI, 23–68%) and 33% (95% CI, 8–58%), respectively (Fig. 1a and b). One patient died a natural death more than 10 years after ASCT in ongoing complete remission. The NRM after ASCT was 0% (Fig. 1c). Thirteen patients relapsed with the original leukemic clone with two patients having late relapses > 2 years after ASCT. The resulting cumulative RI at 5 and 8 years after ASCT was 56% (95% CI, 38–83%) and 67% (95% CI, 47–97%), respectively (Fig. 1d).Fig. 1 Autologous stem cell transplantation in adult standard-risk patients with B- and T-precursor ALL/LBL in first complete remission. a) overall survival b) disease-free survival c) cumulative incidence of non-relapse mortality d) cumulative incidence of relapse
Allogeneic stem cell transplantation in high- and highest-risk patients in first complete remission or relapsed/refractory B- and T-precursor ALL/LBL
The 10-year OS for the entire cohort was 58% (95% CI, 40–76%) (Fig. 2a). The 10-year DFS was 55% (95% CI, 37–73%) (Fig. 2b). The 2- and 5-year NRM was 12% (95% CI, 5–30%) (Fig. 2c). This apparently low NRM most likely resulted from the relatively high proportion of reduced-intensity conditioning used for the transplants (37%). Although NRM was higher in the myeloablative setting (15%; 95% CI, 5–42%) compared with reduced-intensity conditioning (8%; 95% CI, 1–50%), this difference was statistically not significant due to low patient number. Nine patients relapsed within the first 2 years after transplant and the cumulative RI was 30% (95% CI, 17–53%) (Fig. 2d). No relapse occurred > 2 years from transplant.Fig. 2 Allogeneic stem cell transplantation in high-and highest-risk patients in first complete remission or relapsed/refractory B- and T-precursor ALL/LBL. a) overall survival b) disease-free survival c) cumulative incidence of non-relapse mortality d) cumulative incidence of relapse
Twelve (34%) patients developed acute graft-versus-host disease ≥ grade II resulting in a significantly better 5- and 10-year OS (92% vs 42%, p = 0.02) and DFS (79% vs 42%, p = 0.05) mainly due to a trend towards a lower RI compared with patients without acute graft-versus-host disease (13% vs 38%, p = 0.07). There was no difference in NRM between patients with or without acute graft-versus-host disease ≥ grade II (data not shown).
Chronic graft-versus-host disease occurred in 7/32 (22%) patients at risk surviving > 100 days after transplant. There were trends towards a better OS (71% vs 55%, p = 0.33) and a better DFS (71% vs 50%, p = 0.26) due to a lower RI (0% vs 39%, p = 0.06), with no difference in NRM (14% vs 11%) for patients with chronic graft-versus-host disease compared with patients without chronic graft-versus-host disease.
Moreover, no significant differences in outcome and RI of patients with T cell precursor or Ph+ ALL compared with other B-precursor subtypes were observed after allogeneic SCT (data not shown).
Allogeneic stem cell transplantation for relapse after ASCT
The outcome of patients receiving an allogeneic SCT for relapse after ASCT (n=12) was dismal with only two patients surviving > 2 years (OS 11%; 95% CI, 0–38%). It is of note that the majority of patients (8/12; 67%) were refractory to either fludarabine-, clofarabine-, or nelarabine-based salvage regimen prior to allogeneic SCT. As expected, this poor outcome was mainly due to a high 1-year RI of 50% (95% CI, 28–88%) and a high 1-year NRM of 33% (95% CI, 15–74%).
Discussion
This retrospective single center analysis demonstrates that early intensification with ASCT in CR1 in adult patients with standard-risk B- or T-precursor ALL/LBL achieves a long-term OS and DFS of 45% and 33%, respectively, without any treatment-related mortality. Although heterogeneous due to the retrospective nature of our analysis, the distribution of the immunologic subtypes corresponds to that observed within the GMALL studies, and all patients including those with T-LBL received the same induction and intensive consolidation according to the GMALL 07/2003 protocol and the GMALL consensus statement for the treatment of adult patients with T-LBL. Despite the short overall treatment duration with a median time from diagnosis to transplant of 4.9 months, our survival data are comparable to the results of the LALA-94 trial in which standard-risk patients receiving remission induction, consolidation, and a consecutive maintenance program for 2 years achieved a 5-year OS and DFS of 44% and 35%, respectively [14]. In the MRC UKALLXII/ECOG E2993, multinational trial patients were randomized by a donor versus no donor stratification after remission induction and 3 courses of high-dose methotrexate to either receive an allogeneic transplant or, if no donor was available, to receive either consolidation/maintenance or an autograft. Among those patients randomized to chemotherapy versus autograft, the 5-year OS in standard-risk patients was 46% with chemotherapy and only 37% with an autograft [5]. Better results were reported by the PETHEMA ALL-AR-03 trial, in which only high-risk patients were eligible. Patients with good early cytological response and MRD negativity at the end of consolidation were allocated to delayed consolidation/maintenance therapy instead of allogeneic SCT. The 5-year OS and DFS of these MRD negative high-risk patients was 58% and 52%, respectively, particularly emphasizing the importance of highly sensitive MRD measurements for treatment guidance [15].
Whether MRD negativity determined by highly sensitive tools can also be used to significantly reduce the overall treatment duration by early intensification with ASCT in standard-risk patients in CR1 remains an open question. Besides the limited patient number, the fact that NGF for highly sensitive MRD detection was not available in our institution before April 2019 is another major drawback of our retrospective single center analysis [16]. Minimal residual disease negativity by NGF was therefore only documented in 1/1 patient in the autogroup and in 4/5 patients in the allogroup after this diagnostic tool has been implemented. However, although no firm conclusion can be drawn for such small patient numbers, we are convinced that the implementation of NGF will further improve the outcome of our early autotransplant strategy especially by reducing the risk of relapse through an optimized selection of those standard-risk patients who are most likely to benefit from such a short and intensive treatment approach, i.e., MRD negative patients. Furthermore, standard-risk patients who do not achieve MRD negativity (MolCR as defined by ESMO Guidelines) are candidates for one or two early courses of blinatumomab [1]. With blinatumomab, MRD negativity can be achieved in another 70–80% of MRD positive patients as has been demonstrated by the BLAST trial [17]. Whether in blinatumomab responders a consolidation with an autologous transplant with low or even no mortality, as reported here, would be preferable over allogeneic transplantation at least in standard-risk patients because of the high NRM of 37% seen in the BLAST Trial remains an open question [18]. Moreover, it remains elusive whether successful stem cell mobilization is feasible after the administration of blinatumomab to enable subsequent autotransplant at all [19].
Relapse of the underlying disease was the sole reason for the failure of our concept of a short and intensive treatment duration with a final autotransplant. The majority of these relapses were refractory to standard of care salvage chemotherapy thereby resulting in a poor outcome even after subsequent allogeneic stem cell transplantation. Our data in this patient cohort is comparable with the recently published report from the EBMT Acute Leukemia Working Party on the results of allogeneic stem cell transplantation with sequential conditioning using the FLAMSA-RIC strategy in adult patients with refractory or relapsed acute lymphoblastic leukemia with an overall survival of only 17% and a graft-versus-host disease- and leukemia-free survival of only 14% at 2 years [20]. Better results are only obtainable when MRD negativity prior to allotransplant is achieved by implementing targeted therapies [21, 22]. Whether a subsequent allogeneic SCT can be omitted in MRD responders after blinatumomab in the relapsed/refractory setting as it has been supposed by the results of the phase 3 TOWER study has to be interpreted with caution [22].
While the use of ASCT in adult ALL remains a matter of debate, frontline allogeneic SCT is the treatment of choice for Ph1+ and other high-risk B- and T-precursor neoplasms [23–25]. Our results in thirty-five high- and highest-risk and relapsed/refractory patients perfectly fit to the recently published retrospective data of the Acute Leukemia Working Party of the European Society for Blood and Marrow Transplantation (EBMT) showing a long-term survival/cure in about two-thirds of the patients, a relatively low NRM of < 20%, and a RI of about 30% [26]. Whether in the era of potent tyrosine kinase inhibitors (TKIs) in combination with BiTE antibodies or antibody-drug conjugates allogeneic SCT can be replaced by ASCT with less morbidity and mortality at least in those high- and highest-risk patients achieving a deep MolCR remains to be shown by prospective randomized trials.
In conclusion, despite all limitations, our retrospective single center analysis clearly demonstrates that an early autotransplant strategy significantly shortens treatment duration while providing an acceptable long-term outcome in standard-risk patients in first complete remission. The implementation of NGF might be helpful to better define those patients who would benefit most from such a short and intensive treatment approach by reducing the risk of relapse and the consecutive need for a salvage allotransplant with dismal outcome. The increasing availability of MRD evaluation by standardized, highly sensitive diagnostic tools and potent-targeted therapies (e.g., blinatumomab, TKIs) not only allows a more individualized medicine in adult ALL but also asks for the need to reconsider and redefine the role of prolonged chemotherapy versus early autologous or allogeneic stem cell transplantation in the treatment algorithms of various ALL subgroups.
The authors would like to acknowledge all SCT recipients and donors as well as the nursing team for the excellent state of art clinical care given to the patients.
Funding
Open Access funding provided by University of Innsbruck and Medical University of Innsbruck.
Data availability
The datasets generated or analyzed during this study are available from the corresponding author on reasonable request.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflicts of interest.
Ethics approval
Ethical approval was waived by the local Ethics Committee of Innsbruck Medical University of Innsbruck in view of the retrospective nature of the study and all the procedures being performed were part of the routine care.
Consent to participate
Informed consent was obtained from all patients included in the study.
Consent for publication
Patients signed informed consent regarding publishing their data.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | FILGRASTIM | DrugsGivenReaction | CC BY | 33496839 | 18,877,749 | 2021-03 |
What was the outcome of reaction 'Death'? | Early autologous and allogeneic peripheral blood stem cell transplantation for adult patients with acute B and T cell precursor neoplasms: a 12-year single center experience.
Adult acute lymphoblastic leukemia/lymphoma (ALL/LBL) is a rare and heterogeneous malignancy characterized by uncontrolled proliferation of B or T cell precursor cells. Here, we retrospectively analyzed the outcome of early autologous stem cell transplantation in standard-risk patients in first complete remission (n=24) and of allogeneic transplantation in high and highest risk, and relapsed/refractory patients (n=35). The 10-year overall survival after autologous transplantation was 45%. The 10-year overall survival after allogeneic transplantation was 58%. The cumulative incidence of relapse was 29% after allogeneic and 67% after autologous transplantation. The cumulative incidence of non-relapse mortality was 0% after autologous and 12% after allogeneic transplantation. This retrospective single center analysis in a limited number of standard-risk patients clearly demonstrates that early autologous transplantation in first complete remission leads to an acceptable long-term outcome with a short overall treatment duration of less than 6 months compared with more than 2 years with conventional chemotherapy. More sensitive and standardized methods to detect minimal residual disease (MRD) will further help to identify those patients more accurately who are most likely to benefit from such a short and intensive treatment strategy (i.e., MRD negative standard-risk patients) or those who require early targeted therapy (e.g., blinatumomab) in case of MRD positivity. Early allogeneic transplantation results in long-term survival/cure in nearly two-thirds of all high and highest risk, and relapsed/refractory patients.
Introduction
Acute lymphoblastic leukemia/lymphoma (ALL/LBL) is a rare and heterogeneous hematological malignancy. Uncontrolled proliferation of lymphoid progenitor cells committed to the B or T cell lineage in the bone marrow, the peripheral blood, and/or lymphatic and extra-lymphatic tissue characterize the disease. The estimated age- and disease-dependent annual incidence in Europe ranges between 0.17 and 1.45 per 100,000 individuals [1]. Over the past decades, treatment results have improved due to optimized risk stratification, the implementation of monoclonal antibodies (rituximab), BiTE antibodies (blinatumomab), and antibody-drug conjugates (inotuzumab-ozogamicin), as well as the development of highly sensitive diagnostic tools for MRD assessment [2–6]. Besides patient- and disease-related risk-factors and response dynamics, MRD has evolved to the most sensitive marker for the risk of relapse after conventional chemotherapy as well as after stem cell transplantation (SCT) [1, 4, 7]. Allogeneic stem cell transplantation (SCT) is the treatment of choice for high and highest risk as well as relapsed/refractory patients [2, 3]. The conventional treatment approach for standard-risk patients consists of remission induction, consolidation, intensification, and maintenance with an overall treatment duration of 2 or more years [4]. The regimens most commonly used for adults in Europe are based on the pediatric BFM (Berlin-Frankfurt-Münster) protocol, whereas in the USA and many other parts of the world, the hyper-CVAD regimen is preferred [1, 8]. Despite such intensive treatment algorithms, results for adult ALL patients remain unsatisfying [4]. The role of autologous SCT (ASCT) in order to improve treatment results remains controversial. Clinical trials including randomized studies in the pre-MRD era in unselected ALL cohorts have failed to demonstrate any beneficial effect of such a treatment approach [2, 5]. However, studies that are more recent have shown that ASCT might be a suitable option at least for specific subgroups [6, 7].
The main goal of our retrospective single center analysis was to assess whether a significant reduction of treatment duration by early consolidation with ASCT in first complete remission in adult standard-risk ALL is feasible without loss of efficacy compared with the usually recommended conventional post-remission consolidation/maintenance strategy with an overall treatment duration of 2.5–3 years.
Patients and methods
Patients
Between March 2008 and October 2019, fifty-nine adult patients with newly diagnosed B- or T-precursor ALL/LBL received either an autologous (n=24) or allogeneic (n=35) SCT. All patients gave written informed consent. Virtually all patients including those with T-LBL received remission induction (Phase I and II) and consolidation I according to the German Multicenter ALL Protocol for Adults (GMALL 07/2003 Amendment 6, 30.06.2010; Consensus Guidelines of the German Multicenter Study Group for the Treatment of T-lymphoblastic Lymphoma in Adults, Version 1, 18.01.2011) or in patients > 55 years of age according to the GMALL Elderly Protocol 1/2003 (Amendment 5, 21.06.10). Patients with Philadelphia chromosome positive (Ph+) ALL additionally received imatinib 600 mg daily according to the respective protocol.
Characteristics of patients receiving a first ASCT are listed in Table 1. All patients with B- or T-precursor ALL had standard-risk disease as defined by the GMALL 07/2003 protocol. Autologous peripheral blood stem cells were mobilized with granulocyte-colony stimulating factor (G-CSF) 10–20 μg/day starting on day 10 after the first consolidation cycle according to the GMALL 07/2003 protocol. All stem cell products were tested leukemia-free based on conventional fluorescence-activated cell sorter (FACS) analysis, conventional cytogenetics including fluorescence in-situ hybridization (FISH) analysis, and Ig/TCR clonality GeneScan analysis. Median time from diagnosis to transplant in patients in first complete remission (CR1) was 4.9 (range, 2.9–8.9) months. Only one patient received the transplant 20.2 months after diagnosis in second complete remission (CR2) because of a life-threatening infection during consolidation I. Complete remission was determined by bone marrow examination including cytology, conventional FACS analysis, conventional cytogenetics, FISH analysis, real-time polymerase-chain reaction (RT-PCR), Ig/TCR clonality GeneScan analysis, and/or positron emission tomography/computed tomography (PET/CT) scan and/or cerebrospinal fluid assessment when indicated. Additionally, since 2019 next-generation sequencing (NGS) and since April 2019 MRD diagnostics by next-generation flow cytometry (NGF), according to EuroFlow consensus, recommendations (sensitivity < 10−5) were available. According to The GMALL recommendations for stem cell transplantation, all patients received a conditioning regimen with fractionated total body irradiation (fTBI, 12 Gray) combined with either cyclophosphamide (120 mg/m2) or etoposide (60 mg/kg). One patient refusing total body irradiation received intravenous busulfan (12.8 mg/kg) and melphalan (140 mg/m2). All patients with T-LBL (n=7) received the BEAM (carmustine, etoposide, cytarabine, melphalan) regimen [9, 10].Table 1 Characteristics of standard-risk patients receiving an autologous stem cell transplantation in first or second complete remission (n=24)
Median age at the time of diagnosis (years, range) 28 (18–66)
Male/female ratio 16:8
Median time from diagnosis to ASCT in patients in CR1 (months, range) 4.9 (2.9–8.9)
ALL subclassification
B precursor, Ph− 14
T precursor 3
T-LBL 7
Remission status at the time of ASCT
CR1 23
CR2 1
Conditioning regimen
TBI-containing (12 gray) 16
BUMEL 1
BEAM 7
Median CD34 cell number (×106/kg BW, range) 3.74 (1.17–20.87)
Median number of days to leukocyte engraftment (≥ 1.0 G/L) 11 (9–14)
Abbreviations: ALL, acute lymphoblastic leukemia; Ph−, Philadelphia chromosome negative; LBL, lymphoblastic lymphoma; CR1, first complete remission; CR2, second complete remission; TBI, total body irradiation; BUMEL, busulfan, melphalan; BEAM, carmustine, etoposide, cytarabine, melphalan; BW, body weight
Characteristics of patients receiving a first allogeneic SCT are listed in Table 2. All patients receiving an allogeneic SCT in CR1 had high- or highest-risk disease as defined by the GMALL 07/2003 protocol and received remission induction and consolidation according to this protocol (n=29, 83%). One patient was in CR2, and five patients had relapsed/refractory disease. Median time from diagnosis to SCT was 4.5 (range, 1.2–14.8) months. The majority of the patients (83%) had precursor B cell ALL with seven patients being Philadelphia chromosome positive (Ph1+). Conditioning was TBI-containing (fTBI, 12 Gray) and myeloablative in 22/35 (63%) patients, and of fludarabine-based reduced intensity with or without lower doses of fractionated TBI (fTBI, 8 Gray) in the remaining 13/35 (37%) patients. Graft-versus-host disease prophylaxis was calcineurin inhibitor based combined with either methotrexate in the myeloablative setting (n=22) or with mycophenolate mofetil in the reduced-intensity setting (n=13). Eighteen patients (51%) received a graft (mainly peripheral blood stem cells) from an HLA-identical sibling donor. Twelve patients (34%) received allogeneic transplants from HLA-matched unrelated donors (10/10). Five patients (14%) receiving a graft from HLA-mismatched unrelated donors additionally received anti-thymocyte globulin (Grafalon®, 30–60 mg/kg) prior to transplant. All but one patient who died on day+1 because of septic shock engrafted after a median of 13 (range, 8–21) days.Table 2 Characteristics of high- and highest-risk and relapsed/refractory patients receiving an allogeneic stem cell transplantation (n=35)
Median age at the time of diagnosis (years, range) 44 (20–70)
Male/female ratio 13:22
Median time from diagnosis to SCT (months, range) 4.5 (1.2–14.8)
ALL subclassification
B precursor, Ph− 22
B precursor, Ph+ 7
T precursor 6
Remission status at the time of SCT
CR1 29
CR2 1
R/R 5
Conditioning regimen
Myeloablative TBI-containing (12 gray) 22
Reduced-intensity
TBI-containing (8 Gray) 4
Chemotherapy only 9
Stem cell donor
HLA-ID sibling 18
Matched unrelated (10/10) 12
1-Ag mismatched unrelated 5
Median donor age (years, range) 33 (13–66)
Stem cell source
BM 2
PB 33
Recipient/donor sex combination
m/f 7
Others 28
Median CD34 cell number (×106/kg BW, range) 6.15 (2.91–18.9)
Median number of days to leukocyte engraftment (≥ 1.0 G/L) 13 (8–21)
Abbreviations: SCT, stem cell transplantation; ALL, acute lymphoblastic leukemia; Ph−, Philadelphia chromosome negative; Ph+, Philadelphia chromosome positive; CR, complete remission; R/R, relapsed/refractory; TBI, total body irradiation; HLA-ID, human leukocyte antigen-identical; AG, antigen; BM, bone marrow; PB, peripheral blood; m/f, male/female; BW, body weight
Thirteen patients in the autologous group relapsed after a median of 8.2 (range, 2.8–77.1) months, with only two patients having late relapses > 2 years after ASCT. Twelve patients proceeded to an allogeneic SCT within a median of 1.9 (1.3–3.2) months. All but one patient received fludarabine-, clofarabine-, or nelarabine-based salvage therapy prior to transplant [11, 12]. Only four patients achieved a CR2 or CR3. All other patients (8/12, 67%) had refractory disease at the time of allogeneic SCT. Conditioning was of reduced intensity in all patients (fludarabine/busulfan/melphalan (FBM), n=8; thiotepa/busulfan/fludarabine, n=4). All but two patients who died too early due to septic multi-organ failure engrafted after a median of 13 (range 8–23) days.
Study endpoints
The primary endpoint of this retrospective single center analysis was overall survival (OS). Secondary endpoints were disease-free survival (DFS), non-relapse mortality (NRM), and relapse incidence (RI).
Statistical methods
Data were retrospectively reviewed and analyzed as of February 2020. All statistics were computed using NCSS Statistical Software version 19.0.5. The probabilities of OS were calculated using the Kaplan-Meier method from the date of the first transplant until death. DFS was calculated from the date of the first transplant until relapse or death whichever occurred first. The cumulative RI was calculated from the date of the first transplant until relapse with death without relapse as competing risk. The cumulative incidence of NRM was calculated from the date of the first transplant to the date of death without prior relapse with death from relapse as competing risk. NRM for patients receiving an allogeneic SCT because of relapse after a first ASCT was calculated until the date of the allogeneic SCT [13].
Results
Autologous peripheral blood stem cell transplantation in adult standard-risk patients with B- and T-precursor ALL/LBL in first complete remission
The 10-year OS and DFS for the entire cohort was 45% (95% CI, 23–68%) and 33% (95% CI, 8–58%), respectively (Fig. 1a and b). One patient died a natural death more than 10 years after ASCT in ongoing complete remission. The NRM after ASCT was 0% (Fig. 1c). Thirteen patients relapsed with the original leukemic clone with two patients having late relapses > 2 years after ASCT. The resulting cumulative RI at 5 and 8 years after ASCT was 56% (95% CI, 38–83%) and 67% (95% CI, 47–97%), respectively (Fig. 1d).Fig. 1 Autologous stem cell transplantation in adult standard-risk patients with B- and T-precursor ALL/LBL in first complete remission. a) overall survival b) disease-free survival c) cumulative incidence of non-relapse mortality d) cumulative incidence of relapse
Allogeneic stem cell transplantation in high- and highest-risk patients in first complete remission or relapsed/refractory B- and T-precursor ALL/LBL
The 10-year OS for the entire cohort was 58% (95% CI, 40–76%) (Fig. 2a). The 10-year DFS was 55% (95% CI, 37–73%) (Fig. 2b). The 2- and 5-year NRM was 12% (95% CI, 5–30%) (Fig. 2c). This apparently low NRM most likely resulted from the relatively high proportion of reduced-intensity conditioning used for the transplants (37%). Although NRM was higher in the myeloablative setting (15%; 95% CI, 5–42%) compared with reduced-intensity conditioning (8%; 95% CI, 1–50%), this difference was statistically not significant due to low patient number. Nine patients relapsed within the first 2 years after transplant and the cumulative RI was 30% (95% CI, 17–53%) (Fig. 2d). No relapse occurred > 2 years from transplant.Fig. 2 Allogeneic stem cell transplantation in high-and highest-risk patients in first complete remission or relapsed/refractory B- and T-precursor ALL/LBL. a) overall survival b) disease-free survival c) cumulative incidence of non-relapse mortality d) cumulative incidence of relapse
Twelve (34%) patients developed acute graft-versus-host disease ≥ grade II resulting in a significantly better 5- and 10-year OS (92% vs 42%, p = 0.02) and DFS (79% vs 42%, p = 0.05) mainly due to a trend towards a lower RI compared with patients without acute graft-versus-host disease (13% vs 38%, p = 0.07). There was no difference in NRM between patients with or without acute graft-versus-host disease ≥ grade II (data not shown).
Chronic graft-versus-host disease occurred in 7/32 (22%) patients at risk surviving > 100 days after transplant. There were trends towards a better OS (71% vs 55%, p = 0.33) and a better DFS (71% vs 50%, p = 0.26) due to a lower RI (0% vs 39%, p = 0.06), with no difference in NRM (14% vs 11%) for patients with chronic graft-versus-host disease compared with patients without chronic graft-versus-host disease.
Moreover, no significant differences in outcome and RI of patients with T cell precursor or Ph+ ALL compared with other B-precursor subtypes were observed after allogeneic SCT (data not shown).
Allogeneic stem cell transplantation for relapse after ASCT
The outcome of patients receiving an allogeneic SCT for relapse after ASCT (n=12) was dismal with only two patients surviving > 2 years (OS 11%; 95% CI, 0–38%). It is of note that the majority of patients (8/12; 67%) were refractory to either fludarabine-, clofarabine-, or nelarabine-based salvage regimen prior to allogeneic SCT. As expected, this poor outcome was mainly due to a high 1-year RI of 50% (95% CI, 28–88%) and a high 1-year NRM of 33% (95% CI, 15–74%).
Discussion
This retrospective single center analysis demonstrates that early intensification with ASCT in CR1 in adult patients with standard-risk B- or T-precursor ALL/LBL achieves a long-term OS and DFS of 45% and 33%, respectively, without any treatment-related mortality. Although heterogeneous due to the retrospective nature of our analysis, the distribution of the immunologic subtypes corresponds to that observed within the GMALL studies, and all patients including those with T-LBL received the same induction and intensive consolidation according to the GMALL 07/2003 protocol and the GMALL consensus statement for the treatment of adult patients with T-LBL. Despite the short overall treatment duration with a median time from diagnosis to transplant of 4.9 months, our survival data are comparable to the results of the LALA-94 trial in which standard-risk patients receiving remission induction, consolidation, and a consecutive maintenance program for 2 years achieved a 5-year OS and DFS of 44% and 35%, respectively [14]. In the MRC UKALLXII/ECOG E2993, multinational trial patients were randomized by a donor versus no donor stratification after remission induction and 3 courses of high-dose methotrexate to either receive an allogeneic transplant or, if no donor was available, to receive either consolidation/maintenance or an autograft. Among those patients randomized to chemotherapy versus autograft, the 5-year OS in standard-risk patients was 46% with chemotherapy and only 37% with an autograft [5]. Better results were reported by the PETHEMA ALL-AR-03 trial, in which only high-risk patients were eligible. Patients with good early cytological response and MRD negativity at the end of consolidation were allocated to delayed consolidation/maintenance therapy instead of allogeneic SCT. The 5-year OS and DFS of these MRD negative high-risk patients was 58% and 52%, respectively, particularly emphasizing the importance of highly sensitive MRD measurements for treatment guidance [15].
Whether MRD negativity determined by highly sensitive tools can also be used to significantly reduce the overall treatment duration by early intensification with ASCT in standard-risk patients in CR1 remains an open question. Besides the limited patient number, the fact that NGF for highly sensitive MRD detection was not available in our institution before April 2019 is another major drawback of our retrospective single center analysis [16]. Minimal residual disease negativity by NGF was therefore only documented in 1/1 patient in the autogroup and in 4/5 patients in the allogroup after this diagnostic tool has been implemented. However, although no firm conclusion can be drawn for such small patient numbers, we are convinced that the implementation of NGF will further improve the outcome of our early autotransplant strategy especially by reducing the risk of relapse through an optimized selection of those standard-risk patients who are most likely to benefit from such a short and intensive treatment approach, i.e., MRD negative patients. Furthermore, standard-risk patients who do not achieve MRD negativity (MolCR as defined by ESMO Guidelines) are candidates for one or two early courses of blinatumomab [1]. With blinatumomab, MRD negativity can be achieved in another 70–80% of MRD positive patients as has been demonstrated by the BLAST trial [17]. Whether in blinatumomab responders a consolidation with an autologous transplant with low or even no mortality, as reported here, would be preferable over allogeneic transplantation at least in standard-risk patients because of the high NRM of 37% seen in the BLAST Trial remains an open question [18]. Moreover, it remains elusive whether successful stem cell mobilization is feasible after the administration of blinatumomab to enable subsequent autotransplant at all [19].
Relapse of the underlying disease was the sole reason for the failure of our concept of a short and intensive treatment duration with a final autotransplant. The majority of these relapses were refractory to standard of care salvage chemotherapy thereby resulting in a poor outcome even after subsequent allogeneic stem cell transplantation. Our data in this patient cohort is comparable with the recently published report from the EBMT Acute Leukemia Working Party on the results of allogeneic stem cell transplantation with sequential conditioning using the FLAMSA-RIC strategy in adult patients with refractory or relapsed acute lymphoblastic leukemia with an overall survival of only 17% and a graft-versus-host disease- and leukemia-free survival of only 14% at 2 years [20]. Better results are only obtainable when MRD negativity prior to allotransplant is achieved by implementing targeted therapies [21, 22]. Whether a subsequent allogeneic SCT can be omitted in MRD responders after blinatumomab in the relapsed/refractory setting as it has been supposed by the results of the phase 3 TOWER study has to be interpreted with caution [22].
While the use of ASCT in adult ALL remains a matter of debate, frontline allogeneic SCT is the treatment of choice for Ph1+ and other high-risk B- and T-precursor neoplasms [23–25]. Our results in thirty-five high- and highest-risk and relapsed/refractory patients perfectly fit to the recently published retrospective data of the Acute Leukemia Working Party of the European Society for Blood and Marrow Transplantation (EBMT) showing a long-term survival/cure in about two-thirds of the patients, a relatively low NRM of < 20%, and a RI of about 30% [26]. Whether in the era of potent tyrosine kinase inhibitors (TKIs) in combination with BiTE antibodies or antibody-drug conjugates allogeneic SCT can be replaced by ASCT with less morbidity and mortality at least in those high- and highest-risk patients achieving a deep MolCR remains to be shown by prospective randomized trials.
In conclusion, despite all limitations, our retrospective single center analysis clearly demonstrates that an early autotransplant strategy significantly shortens treatment duration while providing an acceptable long-term outcome in standard-risk patients in first complete remission. The implementation of NGF might be helpful to better define those patients who would benefit most from such a short and intensive treatment approach by reducing the risk of relapse and the consecutive need for a salvage allotransplant with dismal outcome. The increasing availability of MRD evaluation by standardized, highly sensitive diagnostic tools and potent-targeted therapies (e.g., blinatumomab, TKIs) not only allows a more individualized medicine in adult ALL but also asks for the need to reconsider and redefine the role of prolonged chemotherapy versus early autologous or allogeneic stem cell transplantation in the treatment algorithms of various ALL subgroups.
The authors would like to acknowledge all SCT recipients and donors as well as the nursing team for the excellent state of art clinical care given to the patients.
Funding
Open Access funding provided by University of Innsbruck and Medical University of Innsbruck.
Data availability
The datasets generated or analyzed during this study are available from the corresponding author on reasonable request.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflicts of interest.
Ethics approval
Ethical approval was waived by the local Ethics Committee of Innsbruck Medical University of Innsbruck in view of the retrospective nature of the study and all the procedures being performed were part of the routine care.
Consent to participate
Informed consent was obtained from all patients included in the study.
Consent for publication
Patients signed informed consent regarding publishing their data.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Fatal | ReactionOutcome | CC BY | 33496839 | 18,877,749 | 2021-03 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Drug ineffective for unapproved indication'. | Tinea Capitis by Microsporum canis in an Elderly Female with Extensive Dermatophyte Infection.
Tinea capitis is a type of dermatophyte infection primarily affecting children. We report a case of an elderly woman with well-controlled diabetes mellitus presenting with a six-month history of erythema with yellow crusts on her scalp and extensive erythematous patches with scales on the body skin. She adopted a stray cat before the disease onset. Dermoscopic findings and manifestation under the Wood's lamp favoured the diagnosis of tinea capitis. Further microscopic examinations of her scalp, including direct KOH and fluorescence stain examination, fungal culture and polymerase chain reaction sequencing identification confirmed the diagnosis of tinea capitis caused by Microsporum canis. Treatment with oral terbinafine was effective. Adult tinea capitis is often misdiagnosed due to its rarity and atypical presentation. However, in some regions, the incidence of tinea capitis in immunocompetent adults is rising which requires the awareness of clinicians. A thorough history (including the animal contacting history), physical examination and further mycological examinations are required for diagnosis. Trichophyton violaceum is the most common dermatophyte species in most regions while adult tinea capitis caused by Microsporum canis is less common. Terbinafine, griseofulvin and itroconazole have been reported to be effective drugs for the treatment of tinea capitis, and terbinafine can be considered as systemic treatment in elderly patients with comorbidities to reduce the drug-drug interaction.
Introduction
Dermatophyte infections are commonly distributed worldwide. According to different clinical manifestations, they are classified into tinea capitis, tinea corporis, tinea cruris, tinea pedis, Majocchi’s granuloma and tinea unguium (dermatophyte onychomycosis). Among them, tinea capitis primarily occurs in children, and extensive tinea corporis occurs mainly in patients with underlying immune disorders such as HIV infection, systematic and topical use of steroids [1, 2]. Here, we report a case of tinea capitis caused by Microsporum canis (M.canis) with extensive superficial dermatophyte infection in an elderly female with well-controlled diabetes mellitus.
Case Report
A 71-year-old woman presented with a six-month history of persistent scalp rashes together with generalized body skin lesions. The rashes first involved her chest skin and then gradually spread to the whole trunk, scalp, groin and all extremities with severe pruritus and malaise. She was diagnosed as seborrhoeic dermatitis and psoriasis previously and treated with compound econazole nitrate and triamcinolone acetonide cream without improvement. She had diabetes mellitus for nearly 5 years. Her fasting blood glucose was controlled within 6–7 mmol/L and postprandial blood glucose within 8–9 mmol/L with subcutaneous insulin. She had been diagnosed with onychomycosis for 2 years without treatment. Physical examination revealed multiple scaly erythematous patches on her scalp with thick greyish-yellow crusts and patches of alopecia with broken hair (Fig. 1a). Bright-green fluorescence was prominent on her hair under Wood’s light illumination (Fig. 1b). Morse code-like hairs, white sheaths and short broken hairs could all be seen under a dermoscope (Fig. 1c). Multiple large erythematous patches could be seen on her face, neck, bilateral ears, trunk, extremities, as well as skin folds including perineum, groin and axillae (Fig. 2a, b). These patches showed elevated and well-demarcated borders, with considerable white superficial scales. All of her toenails revealed discoloration and subungual hyperkeratosis. There was no associated lymphadenopathy.Fig. 1 a Scalp lesions presenting as multiple scaly erythematous patches with thick grayish-yellow crusts. b Broken hair with bright-green fluorescence under the Wood’s light illumination. c Morse code-like hairs (white arrow), white sheath-surrounded hairs (yellow arrow) and short broken hairs (red arrow) under the dermoscope (× 20)
Fig. 2 a, b Extensive erythematous patches on the trunk and arms, with well-demarcated borders and considerable white superficial scales
Direct microscopic examinations (including 10% KOH smear and fluorescence stain) of broken hair revealed ectothrix hyphae and spores (Fig. 3a, b). The direct KOH examination of skin scraping on her trunk and subungual debris were also positive for fungal hyphae. Further history taking revealed contact to a stray cat before the onset of disease. Fungal cultures on Sabouraud dextrose agar (SDA) of both scraping from the patient scalp and the adopted stray cat hair grew Microsporum canis (Fig. 3c, d, e). Unfortunately, fungal cultures of toenails failed to grow any fungal colonies. The sequencing internal transcribed spacer (ITS) rDNA region with ITS1/ITS4 primer further confirmed the strain identification results and the homology of the strain from the patient and cat.Fig. 3 a, b Ectothrix hyphae and spores revealed in the direct KOH examination and fluorescence stain. c, d Gross morphology of M.canis on Sabouraud dextrose agar cultured from the hair of both the patient and the cat (after 10-day incubation at 28 °C). The cottony colony is white from the front and orange from the reverse. e Microscopic examination (× 40) showing septate hyphae and typical macroconidia of M.canis
Considering the coexistence of tinea capitis, extensive tinea corporis, tinea pedis and onychomycosis, the patient was diagnosed as tinea capitis caused by M.canis with extensive dermatophyte infection. She was treated with oral terbinafine 250 mg daily, topical bifonazole solution and ketoconazole cream on her body twice per day. Two weeks later, the rashes on trunk and extremities cleared both clinically and mycologically (Fig. 4a, b). As for her scalp lesion, it took her 6 weeks to be cleared clinically (Fig. 4c, d) and to turn negative mycologically, but 8 weeks for the culture of scrapings and 12 weeks for the result of Wood’s lamp to turn negative, respectively. Her toenails finally got normal gradually in the following 4 months with the continuing treatment of oral terbinafine.Fig. 4 a, b Lesions on the trunk and arms cleared after two-week treatment. c, d Scalp lesions relieved after two-week treatment and cleared after six-week treatment
Discussion
Tinea capitis is a type of fungal infection on the scalp, which primarily affects children aged 3–7 years old [3]. Tinea capitis is uncommon in adults, due to the pH changes and fatty-acids increase in the adult scalp [4]. The proportion of adults among tinea capitis patients was reported to be 2.9% and 4.2% in multi-centre studies from Mexico [5] and Egypt [6], respectively. In single-centre studies, this proportion varies significantly in different countries, from 1.5 to 44.3% [7–11]. In mainland China, adults took up 6.0–13.6% of the tinea capitis population in the 1980s–1990s [12]. This number remained to be 9.0% in the 21th century [13]. In contrast, up to 63% of tinea capitis patients were reported to be adults in a single centre of Taiwan, China [14]. These indicate that adult tinea capitis is becoming less uncommon in some regions, especially in postmenstrual elder women, due to their reduced secretion of fungistatic sebum after menopause. In adult patients, the female-to-male ratio was reported to be 2.2–5.4:1 [4–6, 11, 13, 15] and 26.7–93.5% of the female adults were postmenopausal [10, 11]. Apart from severe immunosuppressive diseases such as HIV infection and post-transplantation, systemic chronic disease including HCV infection (34.4%) and diabetes mellitus (22.4%) has been reported to associate with tinea capitis [6]. Systematic and topical use of corticosteroids are also risk factors [11, 16]. Close contact with animals was reported in 17.1–19.3% of the patients [6, 11, 13]. Furthermore, 20–72% [9, 10, 13] of the adult patients were reported to be immunocompetent. Due to the rise in the occurrence of tinea capitis in immunocompetent adults, adult tinea capitis needs the awareness of clinicians.
Causative agents of adult tinea capitis also vary across geological regions. Trichophyton violaceum was reported to be the main dermatophyte in adults in most countries including Egypt (56.9%, 33/58) [6], Iran (32.0%, 8/25) [10], southern Spain [9] and southern Taiwan, China (74%) [14]. Though M. canis is a major agent causing tinea capitis in children, it is less common in adults [17]. It was reported to account for 15.5% (9/58) infection in Egypt [6] and 16.0% (4/25) in Iran [10]. Exceptionally, M. canis caused 56.5% (13/23) of the adult cases in a single centre in Korea [11]. In mainland China since 2000, Trichophyton violaceum (35.2%, 70/199) was the most common agents, followed by M. canis (21.1%, 42/199), Trichophyton mentagrophyte (16.1%, 32/199) and Trichophyton rubrum (11.5%, 23/199) [13]. Other common dermatophytes isolated in adult patients included Trichophyton tonsurans [5], Trichophyton verrucosum and M. gypseum [6].
Fungal coinfection occurs in 60% of the tinea capitis patients [10]. However, tinea capitis caused by M. canis coexisting with extensive tinea corporis in adults is rare. In this case, the patient had well-controlled diabetes mellitus and denied any other immunosuppression conditions. We hypothesized that M. canis transmitted from animals had stronger pathogenicity which made the skin lesion so generalized. The delay in diagnosis and application of topical corticosteroids also contributed to the lesion generalization. However, the transmission ability of M. canis is weak without animal reservoirs, thus can hardly spread widely among the human population. Its virulence can be lost after about four human-to-human transmissions [3]. This characteristic can be reflected in this case, as the patient did not report any affected family members.
Adult tinea capitis mainly presents as seborrhoeic dermatitis with scaling, grey patch and kerion celsi [11] or pseudo-alopecic plaques [5]. Due to its rarity and atypical presentation, 73.2% of the patients were misdiagnosed during their first consultation in one Korean hospital [11]. Seborrhoeic dermatitis is the diagnosis most commonly confused with tinea capitis in adults [11, 13]. Folliculitis is another major differential diagnosis [4, 11, 13]. The patient in this case was previously misdiagnosed as psoriasis and seborrhoeic dermatitis as these two diseases often present with similar erythema involving scalp and other body areas simultaneously. Thus, a detailed history taking including animal contact and a careful physical examination including searching for broken hair are required. Examinations including Wood’s lamp and dermoscopy can also help in diagnosis and monitoring the treatment effect [18]. Common dermoscopic findings may involve broken hairs, scales, black dots, perifollicular erythema, comma hairs, empty follicles and pustules [11]. In ectothrix infection caused by M.canis, Morse code-like hairs (AKA barcode-like hairs) [19, 20] and white sheaths [21, 22] are typical characteristics under the dermoscope. For a definitive diagnosis, mycological examinations are very important. Among them, 10% KOH microscopic examination of fungal elements is basic for diagnosis and treatment assessment of superficial fungal infection, and novel fluorescent staining can improve the detection rate [23]. Fungal culture can direct the antifungal therapy choice by identification of the dermatophyte.
Treatment of tinea capitis in adults is similar compared with children, yet the age and comorbidities of the patient need to be considered [18]. Terbinafine, griseofulvin and itraconazole have been widely used in the treatment of tinea capitis [10]. As terbinafine hardly influences cytochrome P450 and CYP3A4, it has fewer drug interactions and is increasingly used especially in older patients with multimorbidity [24]. In a meta-analysis comparing the efficacy of griseofulvin and terbinafine in tinea capitis, terbinafine was found to be more effective than griseofulvin for infection by Trichophyton sp., while griseofulvin was shown to be more effective than terbinafine for Microsporum sp. [25]. As M. canis can be resistant to the usual dose of terbinafine (250 mg daily), treatment longer than 4 weeks may be needed for a successful clinical and mycological response [26]. A topical antifungal agent is also advisable for 2-3 months [18]. In this case caused by M.canis, prolonged treatment with oral terbinafine was adopted considering the age and comorbidities of the patient, and the coexisting refractory onychomycosis.
Tinea capitis is not common in adults in most areas. Adult tinea capitis caused by M.canis and co-occurrence with extensive dermatophyte infection are both rarer. This case highlights the challenge in the diagnosis of tinea capitis with extensive dermatophyte infection in adults due to its rarity. In cases presenting as extensive erythema with scales on both scalp and body skin, the fungal infection should be considered. A thorough history including animal contact and physical examination searching for broken hairs are required. Dermoscope, Wood’s lamp examination and further mycological examinations are needed for diagnosis. Terbinafine can be considered as systemic treatment in elderly patients with comorbidities.
Compliance with Ethical Standards
This case report was written complying with the checklist of essential elements instituted by Mycopathologia to guarantee the quality of published case reports [27].
Authors’ Contributions
All the authors involved in the care of the patient. ZY followed up the patient and wrote the text. WC conducted the microscopic examination. ZW and YS involved in the identification of etiologic agent. RL treated the patient and helped to edit the manuscript.
Compliance with Ethical Standards
Conflicts of interest
The authors declare that they have no conflict of interest.
Consent to Participate
Consent from the patient has been obtained and the patient consent form is available.
Consent for Publication
All the authors agree to publish and the written consent for publication has been obtained.
Handling Editor: Vishnu Chaturvedi.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | ECONAZOLE NITRATE, INSULIN NOS, TRIAMCINOLONE ACETONIDE | DrugsGivenReaction | CC BY | 33496917 | 20,204,380 | 2021-05 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Microsporum infection'. | Tinea Capitis by Microsporum canis in an Elderly Female with Extensive Dermatophyte Infection.
Tinea capitis is a type of dermatophyte infection primarily affecting children. We report a case of an elderly woman with well-controlled diabetes mellitus presenting with a six-month history of erythema with yellow crusts on her scalp and extensive erythematous patches with scales on the body skin. She adopted a stray cat before the disease onset. Dermoscopic findings and manifestation under the Wood's lamp favoured the diagnosis of tinea capitis. Further microscopic examinations of her scalp, including direct KOH and fluorescence stain examination, fungal culture and polymerase chain reaction sequencing identification confirmed the diagnosis of tinea capitis caused by Microsporum canis. Treatment with oral terbinafine was effective. Adult tinea capitis is often misdiagnosed due to its rarity and atypical presentation. However, in some regions, the incidence of tinea capitis in immunocompetent adults is rising which requires the awareness of clinicians. A thorough history (including the animal contacting history), physical examination and further mycological examinations are required for diagnosis. Trichophyton violaceum is the most common dermatophyte species in most regions while adult tinea capitis caused by Microsporum canis is less common. Terbinafine, griseofulvin and itroconazole have been reported to be effective drugs for the treatment of tinea capitis, and terbinafine can be considered as systemic treatment in elderly patients with comorbidities to reduce the drug-drug interaction.
Introduction
Dermatophyte infections are commonly distributed worldwide. According to different clinical manifestations, they are classified into tinea capitis, tinea corporis, tinea cruris, tinea pedis, Majocchi’s granuloma and tinea unguium (dermatophyte onychomycosis). Among them, tinea capitis primarily occurs in children, and extensive tinea corporis occurs mainly in patients with underlying immune disorders such as HIV infection, systematic and topical use of steroids [1, 2]. Here, we report a case of tinea capitis caused by Microsporum canis (M.canis) with extensive superficial dermatophyte infection in an elderly female with well-controlled diabetes mellitus.
Case Report
A 71-year-old woman presented with a six-month history of persistent scalp rashes together with generalized body skin lesions. The rashes first involved her chest skin and then gradually spread to the whole trunk, scalp, groin and all extremities with severe pruritus and malaise. She was diagnosed as seborrhoeic dermatitis and psoriasis previously and treated with compound econazole nitrate and triamcinolone acetonide cream without improvement. She had diabetes mellitus for nearly 5 years. Her fasting blood glucose was controlled within 6–7 mmol/L and postprandial blood glucose within 8–9 mmol/L with subcutaneous insulin. She had been diagnosed with onychomycosis for 2 years without treatment. Physical examination revealed multiple scaly erythematous patches on her scalp with thick greyish-yellow crusts and patches of alopecia with broken hair (Fig. 1a). Bright-green fluorescence was prominent on her hair under Wood’s light illumination (Fig. 1b). Morse code-like hairs, white sheaths and short broken hairs could all be seen under a dermoscope (Fig. 1c). Multiple large erythematous patches could be seen on her face, neck, bilateral ears, trunk, extremities, as well as skin folds including perineum, groin and axillae (Fig. 2a, b). These patches showed elevated and well-demarcated borders, with considerable white superficial scales. All of her toenails revealed discoloration and subungual hyperkeratosis. There was no associated lymphadenopathy.Fig. 1 a Scalp lesions presenting as multiple scaly erythematous patches with thick grayish-yellow crusts. b Broken hair with bright-green fluorescence under the Wood’s light illumination. c Morse code-like hairs (white arrow), white sheath-surrounded hairs (yellow arrow) and short broken hairs (red arrow) under the dermoscope (× 20)
Fig. 2 a, b Extensive erythematous patches on the trunk and arms, with well-demarcated borders and considerable white superficial scales
Direct microscopic examinations (including 10% KOH smear and fluorescence stain) of broken hair revealed ectothrix hyphae and spores (Fig. 3a, b). The direct KOH examination of skin scraping on her trunk and subungual debris were also positive for fungal hyphae. Further history taking revealed contact to a stray cat before the onset of disease. Fungal cultures on Sabouraud dextrose agar (SDA) of both scraping from the patient scalp and the adopted stray cat hair grew Microsporum canis (Fig. 3c, d, e). Unfortunately, fungal cultures of toenails failed to grow any fungal colonies. The sequencing internal transcribed spacer (ITS) rDNA region with ITS1/ITS4 primer further confirmed the strain identification results and the homology of the strain from the patient and cat.Fig. 3 a, b Ectothrix hyphae and spores revealed in the direct KOH examination and fluorescence stain. c, d Gross morphology of M.canis on Sabouraud dextrose agar cultured from the hair of both the patient and the cat (after 10-day incubation at 28 °C). The cottony colony is white from the front and orange from the reverse. e Microscopic examination (× 40) showing septate hyphae and typical macroconidia of M.canis
Considering the coexistence of tinea capitis, extensive tinea corporis, tinea pedis and onychomycosis, the patient was diagnosed as tinea capitis caused by M.canis with extensive dermatophyte infection. She was treated with oral terbinafine 250 mg daily, topical bifonazole solution and ketoconazole cream on her body twice per day. Two weeks later, the rashes on trunk and extremities cleared both clinically and mycologically (Fig. 4a, b). As for her scalp lesion, it took her 6 weeks to be cleared clinically (Fig. 4c, d) and to turn negative mycologically, but 8 weeks for the culture of scrapings and 12 weeks for the result of Wood’s lamp to turn negative, respectively. Her toenails finally got normal gradually in the following 4 months with the continuing treatment of oral terbinafine.Fig. 4 a, b Lesions on the trunk and arms cleared after two-week treatment. c, d Scalp lesions relieved after two-week treatment and cleared after six-week treatment
Discussion
Tinea capitis is a type of fungal infection on the scalp, which primarily affects children aged 3–7 years old [3]. Tinea capitis is uncommon in adults, due to the pH changes and fatty-acids increase in the adult scalp [4]. The proportion of adults among tinea capitis patients was reported to be 2.9% and 4.2% in multi-centre studies from Mexico [5] and Egypt [6], respectively. In single-centre studies, this proportion varies significantly in different countries, from 1.5 to 44.3% [7–11]. In mainland China, adults took up 6.0–13.6% of the tinea capitis population in the 1980s–1990s [12]. This number remained to be 9.0% in the 21th century [13]. In contrast, up to 63% of tinea capitis patients were reported to be adults in a single centre of Taiwan, China [14]. These indicate that adult tinea capitis is becoming less uncommon in some regions, especially in postmenstrual elder women, due to their reduced secretion of fungistatic sebum after menopause. In adult patients, the female-to-male ratio was reported to be 2.2–5.4:1 [4–6, 11, 13, 15] and 26.7–93.5% of the female adults were postmenopausal [10, 11]. Apart from severe immunosuppressive diseases such as HIV infection and post-transplantation, systemic chronic disease including HCV infection (34.4%) and diabetes mellitus (22.4%) has been reported to associate with tinea capitis [6]. Systematic and topical use of corticosteroids are also risk factors [11, 16]. Close contact with animals was reported in 17.1–19.3% of the patients [6, 11, 13]. Furthermore, 20–72% [9, 10, 13] of the adult patients were reported to be immunocompetent. Due to the rise in the occurrence of tinea capitis in immunocompetent adults, adult tinea capitis needs the awareness of clinicians.
Causative agents of adult tinea capitis also vary across geological regions. Trichophyton violaceum was reported to be the main dermatophyte in adults in most countries including Egypt (56.9%, 33/58) [6], Iran (32.0%, 8/25) [10], southern Spain [9] and southern Taiwan, China (74%) [14]. Though M. canis is a major agent causing tinea capitis in children, it is less common in adults [17]. It was reported to account for 15.5% (9/58) infection in Egypt [6] and 16.0% (4/25) in Iran [10]. Exceptionally, M. canis caused 56.5% (13/23) of the adult cases in a single centre in Korea [11]. In mainland China since 2000, Trichophyton violaceum (35.2%, 70/199) was the most common agents, followed by M. canis (21.1%, 42/199), Trichophyton mentagrophyte (16.1%, 32/199) and Trichophyton rubrum (11.5%, 23/199) [13]. Other common dermatophytes isolated in adult patients included Trichophyton tonsurans [5], Trichophyton verrucosum and M. gypseum [6].
Fungal coinfection occurs in 60% of the tinea capitis patients [10]. However, tinea capitis caused by M. canis coexisting with extensive tinea corporis in adults is rare. In this case, the patient had well-controlled diabetes mellitus and denied any other immunosuppression conditions. We hypothesized that M. canis transmitted from animals had stronger pathogenicity which made the skin lesion so generalized. The delay in diagnosis and application of topical corticosteroids also contributed to the lesion generalization. However, the transmission ability of M. canis is weak without animal reservoirs, thus can hardly spread widely among the human population. Its virulence can be lost after about four human-to-human transmissions [3]. This characteristic can be reflected in this case, as the patient did not report any affected family members.
Adult tinea capitis mainly presents as seborrhoeic dermatitis with scaling, grey patch and kerion celsi [11] or pseudo-alopecic plaques [5]. Due to its rarity and atypical presentation, 73.2% of the patients were misdiagnosed during their first consultation in one Korean hospital [11]. Seborrhoeic dermatitis is the diagnosis most commonly confused with tinea capitis in adults [11, 13]. Folliculitis is another major differential diagnosis [4, 11, 13]. The patient in this case was previously misdiagnosed as psoriasis and seborrhoeic dermatitis as these two diseases often present with similar erythema involving scalp and other body areas simultaneously. Thus, a detailed history taking including animal contact and a careful physical examination including searching for broken hair are required. Examinations including Wood’s lamp and dermoscopy can also help in diagnosis and monitoring the treatment effect [18]. Common dermoscopic findings may involve broken hairs, scales, black dots, perifollicular erythema, comma hairs, empty follicles and pustules [11]. In ectothrix infection caused by M.canis, Morse code-like hairs (AKA barcode-like hairs) [19, 20] and white sheaths [21, 22] are typical characteristics under the dermoscope. For a definitive diagnosis, mycological examinations are very important. Among them, 10% KOH microscopic examination of fungal elements is basic for diagnosis and treatment assessment of superficial fungal infection, and novel fluorescent staining can improve the detection rate [23]. Fungal culture can direct the antifungal therapy choice by identification of the dermatophyte.
Treatment of tinea capitis in adults is similar compared with children, yet the age and comorbidities of the patient need to be considered [18]. Terbinafine, griseofulvin and itraconazole have been widely used in the treatment of tinea capitis [10]. As terbinafine hardly influences cytochrome P450 and CYP3A4, it has fewer drug interactions and is increasingly used especially in older patients with multimorbidity [24]. In a meta-analysis comparing the efficacy of griseofulvin and terbinafine in tinea capitis, terbinafine was found to be more effective than griseofulvin for infection by Trichophyton sp., while griseofulvin was shown to be more effective than terbinafine for Microsporum sp. [25]. As M. canis can be resistant to the usual dose of terbinafine (250 mg daily), treatment longer than 4 weeks may be needed for a successful clinical and mycological response [26]. A topical antifungal agent is also advisable for 2-3 months [18]. In this case caused by M.canis, prolonged treatment with oral terbinafine was adopted considering the age and comorbidities of the patient, and the coexisting refractory onychomycosis.
Tinea capitis is not common in adults in most areas. Adult tinea capitis caused by M.canis and co-occurrence with extensive dermatophyte infection are both rarer. This case highlights the challenge in the diagnosis of tinea capitis with extensive dermatophyte infection in adults due to its rarity. In cases presenting as extensive erythema with scales on both scalp and body skin, the fungal infection should be considered. A thorough history including animal contact and physical examination searching for broken hairs are required. Dermoscope, Wood’s lamp examination and further mycological examinations are needed for diagnosis. Terbinafine can be considered as systemic treatment in elderly patients with comorbidities.
Compliance with Ethical Standards
This case report was written complying with the checklist of essential elements instituted by Mycopathologia to guarantee the quality of published case reports [27].
Authors’ Contributions
All the authors involved in the care of the patient. ZY followed up the patient and wrote the text. WC conducted the microscopic examination. ZW and YS involved in the identification of etiologic agent. RL treated the patient and helped to edit the manuscript.
Compliance with Ethical Standards
Conflicts of interest
The authors declare that they have no conflict of interest.
Consent to Participate
Consent from the patient has been obtained and the patient consent form is available.
Consent for Publication
All the authors agree to publish and the written consent for publication has been obtained.
Handling Editor: Vishnu Chaturvedi.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | ECONAZOLE NITRATE, INSULIN NOS, TRIAMCINOLONE ACETONIDE | DrugsGivenReaction | CC BY | 33496917 | 20,204,380 | 2021-05 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Off label use'. | Tinea Capitis by Microsporum canis in an Elderly Female with Extensive Dermatophyte Infection.
Tinea capitis is a type of dermatophyte infection primarily affecting children. We report a case of an elderly woman with well-controlled diabetes mellitus presenting with a six-month history of erythema with yellow crusts on her scalp and extensive erythematous patches with scales on the body skin. She adopted a stray cat before the disease onset. Dermoscopic findings and manifestation under the Wood's lamp favoured the diagnosis of tinea capitis. Further microscopic examinations of her scalp, including direct KOH and fluorescence stain examination, fungal culture and polymerase chain reaction sequencing identification confirmed the diagnosis of tinea capitis caused by Microsporum canis. Treatment with oral terbinafine was effective. Adult tinea capitis is often misdiagnosed due to its rarity and atypical presentation. However, in some regions, the incidence of tinea capitis in immunocompetent adults is rising which requires the awareness of clinicians. A thorough history (including the animal contacting history), physical examination and further mycological examinations are required for diagnosis. Trichophyton violaceum is the most common dermatophyte species in most regions while adult tinea capitis caused by Microsporum canis is less common. Terbinafine, griseofulvin and itroconazole have been reported to be effective drugs for the treatment of tinea capitis, and terbinafine can be considered as systemic treatment in elderly patients with comorbidities to reduce the drug-drug interaction.
Introduction
Dermatophyte infections are commonly distributed worldwide. According to different clinical manifestations, they are classified into tinea capitis, tinea corporis, tinea cruris, tinea pedis, Majocchi’s granuloma and tinea unguium (dermatophyte onychomycosis). Among them, tinea capitis primarily occurs in children, and extensive tinea corporis occurs mainly in patients with underlying immune disorders such as HIV infection, systematic and topical use of steroids [1, 2]. Here, we report a case of tinea capitis caused by Microsporum canis (M.canis) with extensive superficial dermatophyte infection in an elderly female with well-controlled diabetes mellitus.
Case Report
A 71-year-old woman presented with a six-month history of persistent scalp rashes together with generalized body skin lesions. The rashes first involved her chest skin and then gradually spread to the whole trunk, scalp, groin and all extremities with severe pruritus and malaise. She was diagnosed as seborrhoeic dermatitis and psoriasis previously and treated with compound econazole nitrate and triamcinolone acetonide cream without improvement. She had diabetes mellitus for nearly 5 years. Her fasting blood glucose was controlled within 6–7 mmol/L and postprandial blood glucose within 8–9 mmol/L with subcutaneous insulin. She had been diagnosed with onychomycosis for 2 years without treatment. Physical examination revealed multiple scaly erythematous patches on her scalp with thick greyish-yellow crusts and patches of alopecia with broken hair (Fig. 1a). Bright-green fluorescence was prominent on her hair under Wood’s light illumination (Fig. 1b). Morse code-like hairs, white sheaths and short broken hairs could all be seen under a dermoscope (Fig. 1c). Multiple large erythematous patches could be seen on her face, neck, bilateral ears, trunk, extremities, as well as skin folds including perineum, groin and axillae (Fig. 2a, b). These patches showed elevated and well-demarcated borders, with considerable white superficial scales. All of her toenails revealed discoloration and subungual hyperkeratosis. There was no associated lymphadenopathy.Fig. 1 a Scalp lesions presenting as multiple scaly erythematous patches with thick grayish-yellow crusts. b Broken hair with bright-green fluorescence under the Wood’s light illumination. c Morse code-like hairs (white arrow), white sheath-surrounded hairs (yellow arrow) and short broken hairs (red arrow) under the dermoscope (× 20)
Fig. 2 a, b Extensive erythematous patches on the trunk and arms, with well-demarcated borders and considerable white superficial scales
Direct microscopic examinations (including 10% KOH smear and fluorescence stain) of broken hair revealed ectothrix hyphae and spores (Fig. 3a, b). The direct KOH examination of skin scraping on her trunk and subungual debris were also positive for fungal hyphae. Further history taking revealed contact to a stray cat before the onset of disease. Fungal cultures on Sabouraud dextrose agar (SDA) of both scraping from the patient scalp and the adopted stray cat hair grew Microsporum canis (Fig. 3c, d, e). Unfortunately, fungal cultures of toenails failed to grow any fungal colonies. The sequencing internal transcribed spacer (ITS) rDNA region with ITS1/ITS4 primer further confirmed the strain identification results and the homology of the strain from the patient and cat.Fig. 3 a, b Ectothrix hyphae and spores revealed in the direct KOH examination and fluorescence stain. c, d Gross morphology of M.canis on Sabouraud dextrose agar cultured from the hair of both the patient and the cat (after 10-day incubation at 28 °C). The cottony colony is white from the front and orange from the reverse. e Microscopic examination (× 40) showing septate hyphae and typical macroconidia of M.canis
Considering the coexistence of tinea capitis, extensive tinea corporis, tinea pedis and onychomycosis, the patient was diagnosed as tinea capitis caused by M.canis with extensive dermatophyte infection. She was treated with oral terbinafine 250 mg daily, topical bifonazole solution and ketoconazole cream on her body twice per day. Two weeks later, the rashes on trunk and extremities cleared both clinically and mycologically (Fig. 4a, b). As for her scalp lesion, it took her 6 weeks to be cleared clinically (Fig. 4c, d) and to turn negative mycologically, but 8 weeks for the culture of scrapings and 12 weeks for the result of Wood’s lamp to turn negative, respectively. Her toenails finally got normal gradually in the following 4 months with the continuing treatment of oral terbinafine.Fig. 4 a, b Lesions on the trunk and arms cleared after two-week treatment. c, d Scalp lesions relieved after two-week treatment and cleared after six-week treatment
Discussion
Tinea capitis is a type of fungal infection on the scalp, which primarily affects children aged 3–7 years old [3]. Tinea capitis is uncommon in adults, due to the pH changes and fatty-acids increase in the adult scalp [4]. The proportion of adults among tinea capitis patients was reported to be 2.9% and 4.2% in multi-centre studies from Mexico [5] and Egypt [6], respectively. In single-centre studies, this proportion varies significantly in different countries, from 1.5 to 44.3% [7–11]. In mainland China, adults took up 6.0–13.6% of the tinea capitis population in the 1980s–1990s [12]. This number remained to be 9.0% in the 21th century [13]. In contrast, up to 63% of tinea capitis patients were reported to be adults in a single centre of Taiwan, China [14]. These indicate that adult tinea capitis is becoming less uncommon in some regions, especially in postmenstrual elder women, due to their reduced secretion of fungistatic sebum after menopause. In adult patients, the female-to-male ratio was reported to be 2.2–5.4:1 [4–6, 11, 13, 15] and 26.7–93.5% of the female adults were postmenopausal [10, 11]. Apart from severe immunosuppressive diseases such as HIV infection and post-transplantation, systemic chronic disease including HCV infection (34.4%) and diabetes mellitus (22.4%) has been reported to associate with tinea capitis [6]. Systematic and topical use of corticosteroids are also risk factors [11, 16]. Close contact with animals was reported in 17.1–19.3% of the patients [6, 11, 13]. Furthermore, 20–72% [9, 10, 13] of the adult patients were reported to be immunocompetent. Due to the rise in the occurrence of tinea capitis in immunocompetent adults, adult tinea capitis needs the awareness of clinicians.
Causative agents of adult tinea capitis also vary across geological regions. Trichophyton violaceum was reported to be the main dermatophyte in adults in most countries including Egypt (56.9%, 33/58) [6], Iran (32.0%, 8/25) [10], southern Spain [9] and southern Taiwan, China (74%) [14]. Though M. canis is a major agent causing tinea capitis in children, it is less common in adults [17]. It was reported to account for 15.5% (9/58) infection in Egypt [6] and 16.0% (4/25) in Iran [10]. Exceptionally, M. canis caused 56.5% (13/23) of the adult cases in a single centre in Korea [11]. In mainland China since 2000, Trichophyton violaceum (35.2%, 70/199) was the most common agents, followed by M. canis (21.1%, 42/199), Trichophyton mentagrophyte (16.1%, 32/199) and Trichophyton rubrum (11.5%, 23/199) [13]. Other common dermatophytes isolated in adult patients included Trichophyton tonsurans [5], Trichophyton verrucosum and M. gypseum [6].
Fungal coinfection occurs in 60% of the tinea capitis patients [10]. However, tinea capitis caused by M. canis coexisting with extensive tinea corporis in adults is rare. In this case, the patient had well-controlled diabetes mellitus and denied any other immunosuppression conditions. We hypothesized that M. canis transmitted from animals had stronger pathogenicity which made the skin lesion so generalized. The delay in diagnosis and application of topical corticosteroids also contributed to the lesion generalization. However, the transmission ability of M. canis is weak without animal reservoirs, thus can hardly spread widely among the human population. Its virulence can be lost after about four human-to-human transmissions [3]. This characteristic can be reflected in this case, as the patient did not report any affected family members.
Adult tinea capitis mainly presents as seborrhoeic dermatitis with scaling, grey patch and kerion celsi [11] or pseudo-alopecic plaques [5]. Due to its rarity and atypical presentation, 73.2% of the patients were misdiagnosed during their first consultation in one Korean hospital [11]. Seborrhoeic dermatitis is the diagnosis most commonly confused with tinea capitis in adults [11, 13]. Folliculitis is another major differential diagnosis [4, 11, 13]. The patient in this case was previously misdiagnosed as psoriasis and seborrhoeic dermatitis as these two diseases often present with similar erythema involving scalp and other body areas simultaneously. Thus, a detailed history taking including animal contact and a careful physical examination including searching for broken hair are required. Examinations including Wood’s lamp and dermoscopy can also help in diagnosis and monitoring the treatment effect [18]. Common dermoscopic findings may involve broken hairs, scales, black dots, perifollicular erythema, comma hairs, empty follicles and pustules [11]. In ectothrix infection caused by M.canis, Morse code-like hairs (AKA barcode-like hairs) [19, 20] and white sheaths [21, 22] are typical characteristics under the dermoscope. For a definitive diagnosis, mycological examinations are very important. Among them, 10% KOH microscopic examination of fungal elements is basic for diagnosis and treatment assessment of superficial fungal infection, and novel fluorescent staining can improve the detection rate [23]. Fungal culture can direct the antifungal therapy choice by identification of the dermatophyte.
Treatment of tinea capitis in adults is similar compared with children, yet the age and comorbidities of the patient need to be considered [18]. Terbinafine, griseofulvin and itraconazole have been widely used in the treatment of tinea capitis [10]. As terbinafine hardly influences cytochrome P450 and CYP3A4, it has fewer drug interactions and is increasingly used especially in older patients with multimorbidity [24]. In a meta-analysis comparing the efficacy of griseofulvin and terbinafine in tinea capitis, terbinafine was found to be more effective than griseofulvin for infection by Trichophyton sp., while griseofulvin was shown to be more effective than terbinafine for Microsporum sp. [25]. As M. canis can be resistant to the usual dose of terbinafine (250 mg daily), treatment longer than 4 weeks may be needed for a successful clinical and mycological response [26]. A topical antifungal agent is also advisable for 2-3 months [18]. In this case caused by M.canis, prolonged treatment with oral terbinafine was adopted considering the age and comorbidities of the patient, and the coexisting refractory onychomycosis.
Tinea capitis is not common in adults in most areas. Adult tinea capitis caused by M.canis and co-occurrence with extensive dermatophyte infection are both rarer. This case highlights the challenge in the diagnosis of tinea capitis with extensive dermatophyte infection in adults due to its rarity. In cases presenting as extensive erythema with scales on both scalp and body skin, the fungal infection should be considered. A thorough history including animal contact and physical examination searching for broken hairs are required. Dermoscope, Wood’s lamp examination and further mycological examinations are needed for diagnosis. Terbinafine can be considered as systemic treatment in elderly patients with comorbidities.
Compliance with Ethical Standards
This case report was written complying with the checklist of essential elements instituted by Mycopathologia to guarantee the quality of published case reports [27].
Authors’ Contributions
All the authors involved in the care of the patient. ZY followed up the patient and wrote the text. WC conducted the microscopic examination. ZW and YS involved in the identification of etiologic agent. RL treated the patient and helped to edit the manuscript.
Compliance with Ethical Standards
Conflicts of interest
The authors declare that they have no conflict of interest.
Consent to Participate
Consent from the patient has been obtained and the patient consent form is available.
Consent for Publication
All the authors agree to publish and the written consent for publication has been obtained.
Handling Editor: Vishnu Chaturvedi.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | ECONAZOLE NITRATE, INSULIN NOS, TRIAMCINOLONE ACETONIDE | DrugsGivenReaction | CC BY | 33496917 | 20,204,380 | 2021-05 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Tinea capitis'. | Tinea Capitis by Microsporum canis in an Elderly Female with Extensive Dermatophyte Infection.
Tinea capitis is a type of dermatophyte infection primarily affecting children. We report a case of an elderly woman with well-controlled diabetes mellitus presenting with a six-month history of erythema with yellow crusts on her scalp and extensive erythematous patches with scales on the body skin. She adopted a stray cat before the disease onset. Dermoscopic findings and manifestation under the Wood's lamp favoured the diagnosis of tinea capitis. Further microscopic examinations of her scalp, including direct KOH and fluorescence stain examination, fungal culture and polymerase chain reaction sequencing identification confirmed the diagnosis of tinea capitis caused by Microsporum canis. Treatment with oral terbinafine was effective. Adult tinea capitis is often misdiagnosed due to its rarity and atypical presentation. However, in some regions, the incidence of tinea capitis in immunocompetent adults is rising which requires the awareness of clinicians. A thorough history (including the animal contacting history), physical examination and further mycological examinations are required for diagnosis. Trichophyton violaceum is the most common dermatophyte species in most regions while adult tinea capitis caused by Microsporum canis is less common. Terbinafine, griseofulvin and itroconazole have been reported to be effective drugs for the treatment of tinea capitis, and terbinafine can be considered as systemic treatment in elderly patients with comorbidities to reduce the drug-drug interaction.
Introduction
Dermatophyte infections are commonly distributed worldwide. According to different clinical manifestations, they are classified into tinea capitis, tinea corporis, tinea cruris, tinea pedis, Majocchi’s granuloma and tinea unguium (dermatophyte onychomycosis). Among them, tinea capitis primarily occurs in children, and extensive tinea corporis occurs mainly in patients with underlying immune disorders such as HIV infection, systematic and topical use of steroids [1, 2]. Here, we report a case of tinea capitis caused by Microsporum canis (M.canis) with extensive superficial dermatophyte infection in an elderly female with well-controlled diabetes mellitus.
Case Report
A 71-year-old woman presented with a six-month history of persistent scalp rashes together with generalized body skin lesions. The rashes first involved her chest skin and then gradually spread to the whole trunk, scalp, groin and all extremities with severe pruritus and malaise. She was diagnosed as seborrhoeic dermatitis and psoriasis previously and treated with compound econazole nitrate and triamcinolone acetonide cream without improvement. She had diabetes mellitus for nearly 5 years. Her fasting blood glucose was controlled within 6–7 mmol/L and postprandial blood glucose within 8–9 mmol/L with subcutaneous insulin. She had been diagnosed with onychomycosis for 2 years without treatment. Physical examination revealed multiple scaly erythematous patches on her scalp with thick greyish-yellow crusts and patches of alopecia with broken hair (Fig. 1a). Bright-green fluorescence was prominent on her hair under Wood’s light illumination (Fig. 1b). Morse code-like hairs, white sheaths and short broken hairs could all be seen under a dermoscope (Fig. 1c). Multiple large erythematous patches could be seen on her face, neck, bilateral ears, trunk, extremities, as well as skin folds including perineum, groin and axillae (Fig. 2a, b). These patches showed elevated and well-demarcated borders, with considerable white superficial scales. All of her toenails revealed discoloration and subungual hyperkeratosis. There was no associated lymphadenopathy.Fig. 1 a Scalp lesions presenting as multiple scaly erythematous patches with thick grayish-yellow crusts. b Broken hair with bright-green fluorescence under the Wood’s light illumination. c Morse code-like hairs (white arrow), white sheath-surrounded hairs (yellow arrow) and short broken hairs (red arrow) under the dermoscope (× 20)
Fig. 2 a, b Extensive erythematous patches on the trunk and arms, with well-demarcated borders and considerable white superficial scales
Direct microscopic examinations (including 10% KOH smear and fluorescence stain) of broken hair revealed ectothrix hyphae and spores (Fig. 3a, b). The direct KOH examination of skin scraping on her trunk and subungual debris were also positive for fungal hyphae. Further history taking revealed contact to a stray cat before the onset of disease. Fungal cultures on Sabouraud dextrose agar (SDA) of both scraping from the patient scalp and the adopted stray cat hair grew Microsporum canis (Fig. 3c, d, e). Unfortunately, fungal cultures of toenails failed to grow any fungal colonies. The sequencing internal transcribed spacer (ITS) rDNA region with ITS1/ITS4 primer further confirmed the strain identification results and the homology of the strain from the patient and cat.Fig. 3 a, b Ectothrix hyphae and spores revealed in the direct KOH examination and fluorescence stain. c, d Gross morphology of M.canis on Sabouraud dextrose agar cultured from the hair of both the patient and the cat (after 10-day incubation at 28 °C). The cottony colony is white from the front and orange from the reverse. e Microscopic examination (× 40) showing septate hyphae and typical macroconidia of M.canis
Considering the coexistence of tinea capitis, extensive tinea corporis, tinea pedis and onychomycosis, the patient was diagnosed as tinea capitis caused by M.canis with extensive dermatophyte infection. She was treated with oral terbinafine 250 mg daily, topical bifonazole solution and ketoconazole cream on her body twice per day. Two weeks later, the rashes on trunk and extremities cleared both clinically and mycologically (Fig. 4a, b). As for her scalp lesion, it took her 6 weeks to be cleared clinically (Fig. 4c, d) and to turn negative mycologically, but 8 weeks for the culture of scrapings and 12 weeks for the result of Wood’s lamp to turn negative, respectively. Her toenails finally got normal gradually in the following 4 months with the continuing treatment of oral terbinafine.Fig. 4 a, b Lesions on the trunk and arms cleared after two-week treatment. c, d Scalp lesions relieved after two-week treatment and cleared after six-week treatment
Discussion
Tinea capitis is a type of fungal infection on the scalp, which primarily affects children aged 3–7 years old [3]. Tinea capitis is uncommon in adults, due to the pH changes and fatty-acids increase in the adult scalp [4]. The proportion of adults among tinea capitis patients was reported to be 2.9% and 4.2% in multi-centre studies from Mexico [5] and Egypt [6], respectively. In single-centre studies, this proportion varies significantly in different countries, from 1.5 to 44.3% [7–11]. In mainland China, adults took up 6.0–13.6% of the tinea capitis population in the 1980s–1990s [12]. This number remained to be 9.0% in the 21th century [13]. In contrast, up to 63% of tinea capitis patients were reported to be adults in a single centre of Taiwan, China [14]. These indicate that adult tinea capitis is becoming less uncommon in some regions, especially in postmenstrual elder women, due to their reduced secretion of fungistatic sebum after menopause. In adult patients, the female-to-male ratio was reported to be 2.2–5.4:1 [4–6, 11, 13, 15] and 26.7–93.5% of the female adults were postmenopausal [10, 11]. Apart from severe immunosuppressive diseases such as HIV infection and post-transplantation, systemic chronic disease including HCV infection (34.4%) and diabetes mellitus (22.4%) has been reported to associate with tinea capitis [6]. Systematic and topical use of corticosteroids are also risk factors [11, 16]. Close contact with animals was reported in 17.1–19.3% of the patients [6, 11, 13]. Furthermore, 20–72% [9, 10, 13] of the adult patients were reported to be immunocompetent. Due to the rise in the occurrence of tinea capitis in immunocompetent adults, adult tinea capitis needs the awareness of clinicians.
Causative agents of adult tinea capitis also vary across geological regions. Trichophyton violaceum was reported to be the main dermatophyte in adults in most countries including Egypt (56.9%, 33/58) [6], Iran (32.0%, 8/25) [10], southern Spain [9] and southern Taiwan, China (74%) [14]. Though M. canis is a major agent causing tinea capitis in children, it is less common in adults [17]. It was reported to account for 15.5% (9/58) infection in Egypt [6] and 16.0% (4/25) in Iran [10]. Exceptionally, M. canis caused 56.5% (13/23) of the adult cases in a single centre in Korea [11]. In mainland China since 2000, Trichophyton violaceum (35.2%, 70/199) was the most common agents, followed by M. canis (21.1%, 42/199), Trichophyton mentagrophyte (16.1%, 32/199) and Trichophyton rubrum (11.5%, 23/199) [13]. Other common dermatophytes isolated in adult patients included Trichophyton tonsurans [5], Trichophyton verrucosum and M. gypseum [6].
Fungal coinfection occurs in 60% of the tinea capitis patients [10]. However, tinea capitis caused by M. canis coexisting with extensive tinea corporis in adults is rare. In this case, the patient had well-controlled diabetes mellitus and denied any other immunosuppression conditions. We hypothesized that M. canis transmitted from animals had stronger pathogenicity which made the skin lesion so generalized. The delay in diagnosis and application of topical corticosteroids also contributed to the lesion generalization. However, the transmission ability of M. canis is weak without animal reservoirs, thus can hardly spread widely among the human population. Its virulence can be lost after about four human-to-human transmissions [3]. This characteristic can be reflected in this case, as the patient did not report any affected family members.
Adult tinea capitis mainly presents as seborrhoeic dermatitis with scaling, grey patch and kerion celsi [11] or pseudo-alopecic plaques [5]. Due to its rarity and atypical presentation, 73.2% of the patients were misdiagnosed during their first consultation in one Korean hospital [11]. Seborrhoeic dermatitis is the diagnosis most commonly confused with tinea capitis in adults [11, 13]. Folliculitis is another major differential diagnosis [4, 11, 13]. The patient in this case was previously misdiagnosed as psoriasis and seborrhoeic dermatitis as these two diseases often present with similar erythema involving scalp and other body areas simultaneously. Thus, a detailed history taking including animal contact and a careful physical examination including searching for broken hair are required. Examinations including Wood’s lamp and dermoscopy can also help in diagnosis and monitoring the treatment effect [18]. Common dermoscopic findings may involve broken hairs, scales, black dots, perifollicular erythema, comma hairs, empty follicles and pustules [11]. In ectothrix infection caused by M.canis, Morse code-like hairs (AKA barcode-like hairs) [19, 20] and white sheaths [21, 22] are typical characteristics under the dermoscope. For a definitive diagnosis, mycological examinations are very important. Among them, 10% KOH microscopic examination of fungal elements is basic for diagnosis and treatment assessment of superficial fungal infection, and novel fluorescent staining can improve the detection rate [23]. Fungal culture can direct the antifungal therapy choice by identification of the dermatophyte.
Treatment of tinea capitis in adults is similar compared with children, yet the age and comorbidities of the patient need to be considered [18]. Terbinafine, griseofulvin and itraconazole have been widely used in the treatment of tinea capitis [10]. As terbinafine hardly influences cytochrome P450 and CYP3A4, it has fewer drug interactions and is increasingly used especially in older patients with multimorbidity [24]. In a meta-analysis comparing the efficacy of griseofulvin and terbinafine in tinea capitis, terbinafine was found to be more effective than griseofulvin for infection by Trichophyton sp., while griseofulvin was shown to be more effective than terbinafine for Microsporum sp. [25]. As M. canis can be resistant to the usual dose of terbinafine (250 mg daily), treatment longer than 4 weeks may be needed for a successful clinical and mycological response [26]. A topical antifungal agent is also advisable for 2-3 months [18]. In this case caused by M.canis, prolonged treatment with oral terbinafine was adopted considering the age and comorbidities of the patient, and the coexisting refractory onychomycosis.
Tinea capitis is not common in adults in most areas. Adult tinea capitis caused by M.canis and co-occurrence with extensive dermatophyte infection are both rarer. This case highlights the challenge in the diagnosis of tinea capitis with extensive dermatophyte infection in adults due to its rarity. In cases presenting as extensive erythema with scales on both scalp and body skin, the fungal infection should be considered. A thorough history including animal contact and physical examination searching for broken hairs are required. Dermoscope, Wood’s lamp examination and further mycological examinations are needed for diagnosis. Terbinafine can be considered as systemic treatment in elderly patients with comorbidities.
Compliance with Ethical Standards
This case report was written complying with the checklist of essential elements instituted by Mycopathologia to guarantee the quality of published case reports [27].
Authors’ Contributions
All the authors involved in the care of the patient. ZY followed up the patient and wrote the text. WC conducted the microscopic examination. ZW and YS involved in the identification of etiologic agent. RL treated the patient and helped to edit the manuscript.
Compliance with Ethical Standards
Conflicts of interest
The authors declare that they have no conflict of interest.
Consent to Participate
Consent from the patient has been obtained and the patient consent form is available.
Consent for Publication
All the authors agree to publish and the written consent for publication has been obtained.
Handling Editor: Vishnu Chaturvedi.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | ECONAZOLE NITRATE, INSULIN NOS, TRIAMCINOLONE ACETONIDE | DrugsGivenReaction | CC BY | 33496917 | 20,204,380 | 2021-05 |
What was the administration route of drug 'TRIAMCINOLONE ACETONIDE'? | Tinea Capitis by Microsporum canis in an Elderly Female with Extensive Dermatophyte Infection.
Tinea capitis is a type of dermatophyte infection primarily affecting children. We report a case of an elderly woman with well-controlled diabetes mellitus presenting with a six-month history of erythema with yellow crusts on her scalp and extensive erythematous patches with scales on the body skin. She adopted a stray cat before the disease onset. Dermoscopic findings and manifestation under the Wood's lamp favoured the diagnosis of tinea capitis. Further microscopic examinations of her scalp, including direct KOH and fluorescence stain examination, fungal culture and polymerase chain reaction sequencing identification confirmed the diagnosis of tinea capitis caused by Microsporum canis. Treatment with oral terbinafine was effective. Adult tinea capitis is often misdiagnosed due to its rarity and atypical presentation. However, in some regions, the incidence of tinea capitis in immunocompetent adults is rising which requires the awareness of clinicians. A thorough history (including the animal contacting history), physical examination and further mycological examinations are required for diagnosis. Trichophyton violaceum is the most common dermatophyte species in most regions while adult tinea capitis caused by Microsporum canis is less common. Terbinafine, griseofulvin and itroconazole have been reported to be effective drugs for the treatment of tinea capitis, and terbinafine can be considered as systemic treatment in elderly patients with comorbidities to reduce the drug-drug interaction.
Introduction
Dermatophyte infections are commonly distributed worldwide. According to different clinical manifestations, they are classified into tinea capitis, tinea corporis, tinea cruris, tinea pedis, Majocchi’s granuloma and tinea unguium (dermatophyte onychomycosis). Among them, tinea capitis primarily occurs in children, and extensive tinea corporis occurs mainly in patients with underlying immune disorders such as HIV infection, systematic and topical use of steroids [1, 2]. Here, we report a case of tinea capitis caused by Microsporum canis (M.canis) with extensive superficial dermatophyte infection in an elderly female with well-controlled diabetes mellitus.
Case Report
A 71-year-old woman presented with a six-month history of persistent scalp rashes together with generalized body skin lesions. The rashes first involved her chest skin and then gradually spread to the whole trunk, scalp, groin and all extremities with severe pruritus and malaise. She was diagnosed as seborrhoeic dermatitis and psoriasis previously and treated with compound econazole nitrate and triamcinolone acetonide cream without improvement. She had diabetes mellitus for nearly 5 years. Her fasting blood glucose was controlled within 6–7 mmol/L and postprandial blood glucose within 8–9 mmol/L with subcutaneous insulin. She had been diagnosed with onychomycosis for 2 years without treatment. Physical examination revealed multiple scaly erythematous patches on her scalp with thick greyish-yellow crusts and patches of alopecia with broken hair (Fig. 1a). Bright-green fluorescence was prominent on her hair under Wood’s light illumination (Fig. 1b). Morse code-like hairs, white sheaths and short broken hairs could all be seen under a dermoscope (Fig. 1c). Multiple large erythematous patches could be seen on her face, neck, bilateral ears, trunk, extremities, as well as skin folds including perineum, groin and axillae (Fig. 2a, b). These patches showed elevated and well-demarcated borders, with considerable white superficial scales. All of her toenails revealed discoloration and subungual hyperkeratosis. There was no associated lymphadenopathy.Fig. 1 a Scalp lesions presenting as multiple scaly erythematous patches with thick grayish-yellow crusts. b Broken hair with bright-green fluorescence under the Wood’s light illumination. c Morse code-like hairs (white arrow), white sheath-surrounded hairs (yellow arrow) and short broken hairs (red arrow) under the dermoscope (× 20)
Fig. 2 a, b Extensive erythematous patches on the trunk and arms, with well-demarcated borders and considerable white superficial scales
Direct microscopic examinations (including 10% KOH smear and fluorescence stain) of broken hair revealed ectothrix hyphae and spores (Fig. 3a, b). The direct KOH examination of skin scraping on her trunk and subungual debris were also positive for fungal hyphae. Further history taking revealed contact to a stray cat before the onset of disease. Fungal cultures on Sabouraud dextrose agar (SDA) of both scraping from the patient scalp and the adopted stray cat hair grew Microsporum canis (Fig. 3c, d, e). Unfortunately, fungal cultures of toenails failed to grow any fungal colonies. The sequencing internal transcribed spacer (ITS) rDNA region with ITS1/ITS4 primer further confirmed the strain identification results and the homology of the strain from the patient and cat.Fig. 3 a, b Ectothrix hyphae and spores revealed in the direct KOH examination and fluorescence stain. c, d Gross morphology of M.canis on Sabouraud dextrose agar cultured from the hair of both the patient and the cat (after 10-day incubation at 28 °C). The cottony colony is white from the front and orange from the reverse. e Microscopic examination (× 40) showing septate hyphae and typical macroconidia of M.canis
Considering the coexistence of tinea capitis, extensive tinea corporis, tinea pedis and onychomycosis, the patient was diagnosed as tinea capitis caused by M.canis with extensive dermatophyte infection. She was treated with oral terbinafine 250 mg daily, topical bifonazole solution and ketoconazole cream on her body twice per day. Two weeks later, the rashes on trunk and extremities cleared both clinically and mycologically (Fig. 4a, b). As for her scalp lesion, it took her 6 weeks to be cleared clinically (Fig. 4c, d) and to turn negative mycologically, but 8 weeks for the culture of scrapings and 12 weeks for the result of Wood’s lamp to turn negative, respectively. Her toenails finally got normal gradually in the following 4 months with the continuing treatment of oral terbinafine.Fig. 4 a, b Lesions on the trunk and arms cleared after two-week treatment. c, d Scalp lesions relieved after two-week treatment and cleared after six-week treatment
Discussion
Tinea capitis is a type of fungal infection on the scalp, which primarily affects children aged 3–7 years old [3]. Tinea capitis is uncommon in adults, due to the pH changes and fatty-acids increase in the adult scalp [4]. The proportion of adults among tinea capitis patients was reported to be 2.9% and 4.2% in multi-centre studies from Mexico [5] and Egypt [6], respectively. In single-centre studies, this proportion varies significantly in different countries, from 1.5 to 44.3% [7–11]. In mainland China, adults took up 6.0–13.6% of the tinea capitis population in the 1980s–1990s [12]. This number remained to be 9.0% in the 21th century [13]. In contrast, up to 63% of tinea capitis patients were reported to be adults in a single centre of Taiwan, China [14]. These indicate that adult tinea capitis is becoming less uncommon in some regions, especially in postmenstrual elder women, due to their reduced secretion of fungistatic sebum after menopause. In adult patients, the female-to-male ratio was reported to be 2.2–5.4:1 [4–6, 11, 13, 15] and 26.7–93.5% of the female adults were postmenopausal [10, 11]. Apart from severe immunosuppressive diseases such as HIV infection and post-transplantation, systemic chronic disease including HCV infection (34.4%) and diabetes mellitus (22.4%) has been reported to associate with tinea capitis [6]. Systematic and topical use of corticosteroids are also risk factors [11, 16]. Close contact with animals was reported in 17.1–19.3% of the patients [6, 11, 13]. Furthermore, 20–72% [9, 10, 13] of the adult patients were reported to be immunocompetent. Due to the rise in the occurrence of tinea capitis in immunocompetent adults, adult tinea capitis needs the awareness of clinicians.
Causative agents of adult tinea capitis also vary across geological regions. Trichophyton violaceum was reported to be the main dermatophyte in adults in most countries including Egypt (56.9%, 33/58) [6], Iran (32.0%, 8/25) [10], southern Spain [9] and southern Taiwan, China (74%) [14]. Though M. canis is a major agent causing tinea capitis in children, it is less common in adults [17]. It was reported to account for 15.5% (9/58) infection in Egypt [6] and 16.0% (4/25) in Iran [10]. Exceptionally, M. canis caused 56.5% (13/23) of the adult cases in a single centre in Korea [11]. In mainland China since 2000, Trichophyton violaceum (35.2%, 70/199) was the most common agents, followed by M. canis (21.1%, 42/199), Trichophyton mentagrophyte (16.1%, 32/199) and Trichophyton rubrum (11.5%, 23/199) [13]. Other common dermatophytes isolated in adult patients included Trichophyton tonsurans [5], Trichophyton verrucosum and M. gypseum [6].
Fungal coinfection occurs in 60% of the tinea capitis patients [10]. However, tinea capitis caused by M. canis coexisting with extensive tinea corporis in adults is rare. In this case, the patient had well-controlled diabetes mellitus and denied any other immunosuppression conditions. We hypothesized that M. canis transmitted from animals had stronger pathogenicity which made the skin lesion so generalized. The delay in diagnosis and application of topical corticosteroids also contributed to the lesion generalization. However, the transmission ability of M. canis is weak without animal reservoirs, thus can hardly spread widely among the human population. Its virulence can be lost after about four human-to-human transmissions [3]. This characteristic can be reflected in this case, as the patient did not report any affected family members.
Adult tinea capitis mainly presents as seborrhoeic dermatitis with scaling, grey patch and kerion celsi [11] or pseudo-alopecic plaques [5]. Due to its rarity and atypical presentation, 73.2% of the patients were misdiagnosed during their first consultation in one Korean hospital [11]. Seborrhoeic dermatitis is the diagnosis most commonly confused with tinea capitis in adults [11, 13]. Folliculitis is another major differential diagnosis [4, 11, 13]. The patient in this case was previously misdiagnosed as psoriasis and seborrhoeic dermatitis as these two diseases often present with similar erythema involving scalp and other body areas simultaneously. Thus, a detailed history taking including animal contact and a careful physical examination including searching for broken hair are required. Examinations including Wood’s lamp and dermoscopy can also help in diagnosis and monitoring the treatment effect [18]. Common dermoscopic findings may involve broken hairs, scales, black dots, perifollicular erythema, comma hairs, empty follicles and pustules [11]. In ectothrix infection caused by M.canis, Morse code-like hairs (AKA barcode-like hairs) [19, 20] and white sheaths [21, 22] are typical characteristics under the dermoscope. For a definitive diagnosis, mycological examinations are very important. Among them, 10% KOH microscopic examination of fungal elements is basic for diagnosis and treatment assessment of superficial fungal infection, and novel fluorescent staining can improve the detection rate [23]. Fungal culture can direct the antifungal therapy choice by identification of the dermatophyte.
Treatment of tinea capitis in adults is similar compared with children, yet the age and comorbidities of the patient need to be considered [18]. Terbinafine, griseofulvin and itraconazole have been widely used in the treatment of tinea capitis [10]. As terbinafine hardly influences cytochrome P450 and CYP3A4, it has fewer drug interactions and is increasingly used especially in older patients with multimorbidity [24]. In a meta-analysis comparing the efficacy of griseofulvin and terbinafine in tinea capitis, terbinafine was found to be more effective than griseofulvin for infection by Trichophyton sp., while griseofulvin was shown to be more effective than terbinafine for Microsporum sp. [25]. As M. canis can be resistant to the usual dose of terbinafine (250 mg daily), treatment longer than 4 weeks may be needed for a successful clinical and mycological response [26]. A topical antifungal agent is also advisable for 2-3 months [18]. In this case caused by M.canis, prolonged treatment with oral terbinafine was adopted considering the age and comorbidities of the patient, and the coexisting refractory onychomycosis.
Tinea capitis is not common in adults in most areas. Adult tinea capitis caused by M.canis and co-occurrence with extensive dermatophyte infection are both rarer. This case highlights the challenge in the diagnosis of tinea capitis with extensive dermatophyte infection in adults due to its rarity. In cases presenting as extensive erythema with scales on both scalp and body skin, the fungal infection should be considered. A thorough history including animal contact and physical examination searching for broken hairs are required. Dermoscope, Wood’s lamp examination and further mycological examinations are needed for diagnosis. Terbinafine can be considered as systemic treatment in elderly patients with comorbidities.
Compliance with Ethical Standards
This case report was written complying with the checklist of essential elements instituted by Mycopathologia to guarantee the quality of published case reports [27].
Authors’ Contributions
All the authors involved in the care of the patient. ZY followed up the patient and wrote the text. WC conducted the microscopic examination. ZW and YS involved in the identification of etiologic agent. RL treated the patient and helped to edit the manuscript.
Compliance with Ethical Standards
Conflicts of interest
The authors declare that they have no conflict of interest.
Consent to Participate
Consent from the patient has been obtained and the patient consent form is available.
Consent for Publication
All the authors agree to publish and the written consent for publication has been obtained.
Handling Editor: Vishnu Chaturvedi.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Topical | DrugAdministrationRoute | CC BY | 33496917 | 20,204,380 | 2021-05 |
What was the outcome of reaction 'Microsporum infection'? | Tinea Capitis by Microsporum canis in an Elderly Female with Extensive Dermatophyte Infection.
Tinea capitis is a type of dermatophyte infection primarily affecting children. We report a case of an elderly woman with well-controlled diabetes mellitus presenting with a six-month history of erythema with yellow crusts on her scalp and extensive erythematous patches with scales on the body skin. She adopted a stray cat before the disease onset. Dermoscopic findings and manifestation under the Wood's lamp favoured the diagnosis of tinea capitis. Further microscopic examinations of her scalp, including direct KOH and fluorescence stain examination, fungal culture and polymerase chain reaction sequencing identification confirmed the diagnosis of tinea capitis caused by Microsporum canis. Treatment with oral terbinafine was effective. Adult tinea capitis is often misdiagnosed due to its rarity and atypical presentation. However, in some regions, the incidence of tinea capitis in immunocompetent adults is rising which requires the awareness of clinicians. A thorough history (including the animal contacting history), physical examination and further mycological examinations are required for diagnosis. Trichophyton violaceum is the most common dermatophyte species in most regions while adult tinea capitis caused by Microsporum canis is less common. Terbinafine, griseofulvin and itroconazole have been reported to be effective drugs for the treatment of tinea capitis, and terbinafine can be considered as systemic treatment in elderly patients with comorbidities to reduce the drug-drug interaction.
Introduction
Dermatophyte infections are commonly distributed worldwide. According to different clinical manifestations, they are classified into tinea capitis, tinea corporis, tinea cruris, tinea pedis, Majocchi’s granuloma and tinea unguium (dermatophyte onychomycosis). Among them, tinea capitis primarily occurs in children, and extensive tinea corporis occurs mainly in patients with underlying immune disorders such as HIV infection, systematic and topical use of steroids [1, 2]. Here, we report a case of tinea capitis caused by Microsporum canis (M.canis) with extensive superficial dermatophyte infection in an elderly female with well-controlled diabetes mellitus.
Case Report
A 71-year-old woman presented with a six-month history of persistent scalp rashes together with generalized body skin lesions. The rashes first involved her chest skin and then gradually spread to the whole trunk, scalp, groin and all extremities with severe pruritus and malaise. She was diagnosed as seborrhoeic dermatitis and psoriasis previously and treated with compound econazole nitrate and triamcinolone acetonide cream without improvement. She had diabetes mellitus for nearly 5 years. Her fasting blood glucose was controlled within 6–7 mmol/L and postprandial blood glucose within 8–9 mmol/L with subcutaneous insulin. She had been diagnosed with onychomycosis for 2 years without treatment. Physical examination revealed multiple scaly erythematous patches on her scalp with thick greyish-yellow crusts and patches of alopecia with broken hair (Fig. 1a). Bright-green fluorescence was prominent on her hair under Wood’s light illumination (Fig. 1b). Morse code-like hairs, white sheaths and short broken hairs could all be seen under a dermoscope (Fig. 1c). Multiple large erythematous patches could be seen on her face, neck, bilateral ears, trunk, extremities, as well as skin folds including perineum, groin and axillae (Fig. 2a, b). These patches showed elevated and well-demarcated borders, with considerable white superficial scales. All of her toenails revealed discoloration and subungual hyperkeratosis. There was no associated lymphadenopathy.Fig. 1 a Scalp lesions presenting as multiple scaly erythematous patches with thick grayish-yellow crusts. b Broken hair with bright-green fluorescence under the Wood’s light illumination. c Morse code-like hairs (white arrow), white sheath-surrounded hairs (yellow arrow) and short broken hairs (red arrow) under the dermoscope (× 20)
Fig. 2 a, b Extensive erythematous patches on the trunk and arms, with well-demarcated borders and considerable white superficial scales
Direct microscopic examinations (including 10% KOH smear and fluorescence stain) of broken hair revealed ectothrix hyphae and spores (Fig. 3a, b). The direct KOH examination of skin scraping on her trunk and subungual debris were also positive for fungal hyphae. Further history taking revealed contact to a stray cat before the onset of disease. Fungal cultures on Sabouraud dextrose agar (SDA) of both scraping from the patient scalp and the adopted stray cat hair grew Microsporum canis (Fig. 3c, d, e). Unfortunately, fungal cultures of toenails failed to grow any fungal colonies. The sequencing internal transcribed spacer (ITS) rDNA region with ITS1/ITS4 primer further confirmed the strain identification results and the homology of the strain from the patient and cat.Fig. 3 a, b Ectothrix hyphae and spores revealed in the direct KOH examination and fluorescence stain. c, d Gross morphology of M.canis on Sabouraud dextrose agar cultured from the hair of both the patient and the cat (after 10-day incubation at 28 °C). The cottony colony is white from the front and orange from the reverse. e Microscopic examination (× 40) showing septate hyphae and typical macroconidia of M.canis
Considering the coexistence of tinea capitis, extensive tinea corporis, tinea pedis and onychomycosis, the patient was diagnosed as tinea capitis caused by M.canis with extensive dermatophyte infection. She was treated with oral terbinafine 250 mg daily, topical bifonazole solution and ketoconazole cream on her body twice per day. Two weeks later, the rashes on trunk and extremities cleared both clinically and mycologically (Fig. 4a, b). As for her scalp lesion, it took her 6 weeks to be cleared clinically (Fig. 4c, d) and to turn negative mycologically, but 8 weeks for the culture of scrapings and 12 weeks for the result of Wood’s lamp to turn negative, respectively. Her toenails finally got normal gradually in the following 4 months with the continuing treatment of oral terbinafine.Fig. 4 a, b Lesions on the trunk and arms cleared after two-week treatment. c, d Scalp lesions relieved after two-week treatment and cleared after six-week treatment
Discussion
Tinea capitis is a type of fungal infection on the scalp, which primarily affects children aged 3–7 years old [3]. Tinea capitis is uncommon in adults, due to the pH changes and fatty-acids increase in the adult scalp [4]. The proportion of adults among tinea capitis patients was reported to be 2.9% and 4.2% in multi-centre studies from Mexico [5] and Egypt [6], respectively. In single-centre studies, this proportion varies significantly in different countries, from 1.5 to 44.3% [7–11]. In mainland China, adults took up 6.0–13.6% of the tinea capitis population in the 1980s–1990s [12]. This number remained to be 9.0% in the 21th century [13]. In contrast, up to 63% of tinea capitis patients were reported to be adults in a single centre of Taiwan, China [14]. These indicate that adult tinea capitis is becoming less uncommon in some regions, especially in postmenstrual elder women, due to their reduced secretion of fungistatic sebum after menopause. In adult patients, the female-to-male ratio was reported to be 2.2–5.4:1 [4–6, 11, 13, 15] and 26.7–93.5% of the female adults were postmenopausal [10, 11]. Apart from severe immunosuppressive diseases such as HIV infection and post-transplantation, systemic chronic disease including HCV infection (34.4%) and diabetes mellitus (22.4%) has been reported to associate with tinea capitis [6]. Systematic and topical use of corticosteroids are also risk factors [11, 16]. Close contact with animals was reported in 17.1–19.3% of the patients [6, 11, 13]. Furthermore, 20–72% [9, 10, 13] of the adult patients were reported to be immunocompetent. Due to the rise in the occurrence of tinea capitis in immunocompetent adults, adult tinea capitis needs the awareness of clinicians.
Causative agents of adult tinea capitis also vary across geological regions. Trichophyton violaceum was reported to be the main dermatophyte in adults in most countries including Egypt (56.9%, 33/58) [6], Iran (32.0%, 8/25) [10], southern Spain [9] and southern Taiwan, China (74%) [14]. Though M. canis is a major agent causing tinea capitis in children, it is less common in adults [17]. It was reported to account for 15.5% (9/58) infection in Egypt [6] and 16.0% (4/25) in Iran [10]. Exceptionally, M. canis caused 56.5% (13/23) of the adult cases in a single centre in Korea [11]. In mainland China since 2000, Trichophyton violaceum (35.2%, 70/199) was the most common agents, followed by M. canis (21.1%, 42/199), Trichophyton mentagrophyte (16.1%, 32/199) and Trichophyton rubrum (11.5%, 23/199) [13]. Other common dermatophytes isolated in adult patients included Trichophyton tonsurans [5], Trichophyton verrucosum and M. gypseum [6].
Fungal coinfection occurs in 60% of the tinea capitis patients [10]. However, tinea capitis caused by M. canis coexisting with extensive tinea corporis in adults is rare. In this case, the patient had well-controlled diabetes mellitus and denied any other immunosuppression conditions. We hypothesized that M. canis transmitted from animals had stronger pathogenicity which made the skin lesion so generalized. The delay in diagnosis and application of topical corticosteroids also contributed to the lesion generalization. However, the transmission ability of M. canis is weak without animal reservoirs, thus can hardly spread widely among the human population. Its virulence can be lost after about four human-to-human transmissions [3]. This characteristic can be reflected in this case, as the patient did not report any affected family members.
Adult tinea capitis mainly presents as seborrhoeic dermatitis with scaling, grey patch and kerion celsi [11] or pseudo-alopecic plaques [5]. Due to its rarity and atypical presentation, 73.2% of the patients were misdiagnosed during their first consultation in one Korean hospital [11]. Seborrhoeic dermatitis is the diagnosis most commonly confused with tinea capitis in adults [11, 13]. Folliculitis is another major differential diagnosis [4, 11, 13]. The patient in this case was previously misdiagnosed as psoriasis and seborrhoeic dermatitis as these two diseases often present with similar erythema involving scalp and other body areas simultaneously. Thus, a detailed history taking including animal contact and a careful physical examination including searching for broken hair are required. Examinations including Wood’s lamp and dermoscopy can also help in diagnosis and monitoring the treatment effect [18]. Common dermoscopic findings may involve broken hairs, scales, black dots, perifollicular erythema, comma hairs, empty follicles and pustules [11]. In ectothrix infection caused by M.canis, Morse code-like hairs (AKA barcode-like hairs) [19, 20] and white sheaths [21, 22] are typical characteristics under the dermoscope. For a definitive diagnosis, mycological examinations are very important. Among them, 10% KOH microscopic examination of fungal elements is basic for diagnosis and treatment assessment of superficial fungal infection, and novel fluorescent staining can improve the detection rate [23]. Fungal culture can direct the antifungal therapy choice by identification of the dermatophyte.
Treatment of tinea capitis in adults is similar compared with children, yet the age and comorbidities of the patient need to be considered [18]. Terbinafine, griseofulvin and itraconazole have been widely used in the treatment of tinea capitis [10]. As terbinafine hardly influences cytochrome P450 and CYP3A4, it has fewer drug interactions and is increasingly used especially in older patients with multimorbidity [24]. In a meta-analysis comparing the efficacy of griseofulvin and terbinafine in tinea capitis, terbinafine was found to be more effective than griseofulvin for infection by Trichophyton sp., while griseofulvin was shown to be more effective than terbinafine for Microsporum sp. [25]. As M. canis can be resistant to the usual dose of terbinafine (250 mg daily), treatment longer than 4 weeks may be needed for a successful clinical and mycological response [26]. A topical antifungal agent is also advisable for 2-3 months [18]. In this case caused by M.canis, prolonged treatment with oral terbinafine was adopted considering the age and comorbidities of the patient, and the coexisting refractory onychomycosis.
Tinea capitis is not common in adults in most areas. Adult tinea capitis caused by M.canis and co-occurrence with extensive dermatophyte infection are both rarer. This case highlights the challenge in the diagnosis of tinea capitis with extensive dermatophyte infection in adults due to its rarity. In cases presenting as extensive erythema with scales on both scalp and body skin, the fungal infection should be considered. A thorough history including animal contact and physical examination searching for broken hairs are required. Dermoscope, Wood’s lamp examination and further mycological examinations are needed for diagnosis. Terbinafine can be considered as systemic treatment in elderly patients with comorbidities.
Compliance with Ethical Standards
This case report was written complying with the checklist of essential elements instituted by Mycopathologia to guarantee the quality of published case reports [27].
Authors’ Contributions
All the authors involved in the care of the patient. ZY followed up the patient and wrote the text. WC conducted the microscopic examination. ZW and YS involved in the identification of etiologic agent. RL treated the patient and helped to edit the manuscript.
Compliance with Ethical Standards
Conflicts of interest
The authors declare that they have no conflict of interest.
Consent to Participate
Consent from the patient has been obtained and the patient consent form is available.
Consent for Publication
All the authors agree to publish and the written consent for publication has been obtained.
Handling Editor: Vishnu Chaturvedi.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Recovering | ReactionOutcome | CC BY | 33496917 | 20,204,380 | 2021-05 |
What was the outcome of reaction 'Tinea capitis'? | Tinea Capitis by Microsporum canis in an Elderly Female with Extensive Dermatophyte Infection.
Tinea capitis is a type of dermatophyte infection primarily affecting children. We report a case of an elderly woman with well-controlled diabetes mellitus presenting with a six-month history of erythema with yellow crusts on her scalp and extensive erythematous patches with scales on the body skin. She adopted a stray cat before the disease onset. Dermoscopic findings and manifestation under the Wood's lamp favoured the diagnosis of tinea capitis. Further microscopic examinations of her scalp, including direct KOH and fluorescence stain examination, fungal culture and polymerase chain reaction sequencing identification confirmed the diagnosis of tinea capitis caused by Microsporum canis. Treatment with oral terbinafine was effective. Adult tinea capitis is often misdiagnosed due to its rarity and atypical presentation. However, in some regions, the incidence of tinea capitis in immunocompetent adults is rising which requires the awareness of clinicians. A thorough history (including the animal contacting history), physical examination and further mycological examinations are required for diagnosis. Trichophyton violaceum is the most common dermatophyte species in most regions while adult tinea capitis caused by Microsporum canis is less common. Terbinafine, griseofulvin and itroconazole have been reported to be effective drugs for the treatment of tinea capitis, and terbinafine can be considered as systemic treatment in elderly patients with comorbidities to reduce the drug-drug interaction.
Introduction
Dermatophyte infections are commonly distributed worldwide. According to different clinical manifestations, they are classified into tinea capitis, tinea corporis, tinea cruris, tinea pedis, Majocchi’s granuloma and tinea unguium (dermatophyte onychomycosis). Among them, tinea capitis primarily occurs in children, and extensive tinea corporis occurs mainly in patients with underlying immune disorders such as HIV infection, systematic and topical use of steroids [1, 2]. Here, we report a case of tinea capitis caused by Microsporum canis (M.canis) with extensive superficial dermatophyte infection in an elderly female with well-controlled diabetes mellitus.
Case Report
A 71-year-old woman presented with a six-month history of persistent scalp rashes together with generalized body skin lesions. The rashes first involved her chest skin and then gradually spread to the whole trunk, scalp, groin and all extremities with severe pruritus and malaise. She was diagnosed as seborrhoeic dermatitis and psoriasis previously and treated with compound econazole nitrate and triamcinolone acetonide cream without improvement. She had diabetes mellitus for nearly 5 years. Her fasting blood glucose was controlled within 6–7 mmol/L and postprandial blood glucose within 8–9 mmol/L with subcutaneous insulin. She had been diagnosed with onychomycosis for 2 years without treatment. Physical examination revealed multiple scaly erythematous patches on her scalp with thick greyish-yellow crusts and patches of alopecia with broken hair (Fig. 1a). Bright-green fluorescence was prominent on her hair under Wood’s light illumination (Fig. 1b). Morse code-like hairs, white sheaths and short broken hairs could all be seen under a dermoscope (Fig. 1c). Multiple large erythematous patches could be seen on her face, neck, bilateral ears, trunk, extremities, as well as skin folds including perineum, groin and axillae (Fig. 2a, b). These patches showed elevated and well-demarcated borders, with considerable white superficial scales. All of her toenails revealed discoloration and subungual hyperkeratosis. There was no associated lymphadenopathy.Fig. 1 a Scalp lesions presenting as multiple scaly erythematous patches with thick grayish-yellow crusts. b Broken hair with bright-green fluorescence under the Wood’s light illumination. c Morse code-like hairs (white arrow), white sheath-surrounded hairs (yellow arrow) and short broken hairs (red arrow) under the dermoscope (× 20)
Fig. 2 a, b Extensive erythematous patches on the trunk and arms, with well-demarcated borders and considerable white superficial scales
Direct microscopic examinations (including 10% KOH smear and fluorescence stain) of broken hair revealed ectothrix hyphae and spores (Fig. 3a, b). The direct KOH examination of skin scraping on her trunk and subungual debris were also positive for fungal hyphae. Further history taking revealed contact to a stray cat before the onset of disease. Fungal cultures on Sabouraud dextrose agar (SDA) of both scraping from the patient scalp and the adopted stray cat hair grew Microsporum canis (Fig. 3c, d, e). Unfortunately, fungal cultures of toenails failed to grow any fungal colonies. The sequencing internal transcribed spacer (ITS) rDNA region with ITS1/ITS4 primer further confirmed the strain identification results and the homology of the strain from the patient and cat.Fig. 3 a, b Ectothrix hyphae and spores revealed in the direct KOH examination and fluorescence stain. c, d Gross morphology of M.canis on Sabouraud dextrose agar cultured from the hair of both the patient and the cat (after 10-day incubation at 28 °C). The cottony colony is white from the front and orange from the reverse. e Microscopic examination (× 40) showing septate hyphae and typical macroconidia of M.canis
Considering the coexistence of tinea capitis, extensive tinea corporis, tinea pedis and onychomycosis, the patient was diagnosed as tinea capitis caused by M.canis with extensive dermatophyte infection. She was treated with oral terbinafine 250 mg daily, topical bifonazole solution and ketoconazole cream on her body twice per day. Two weeks later, the rashes on trunk and extremities cleared both clinically and mycologically (Fig. 4a, b). As for her scalp lesion, it took her 6 weeks to be cleared clinically (Fig. 4c, d) and to turn negative mycologically, but 8 weeks for the culture of scrapings and 12 weeks for the result of Wood’s lamp to turn negative, respectively. Her toenails finally got normal gradually in the following 4 months with the continuing treatment of oral terbinafine.Fig. 4 a, b Lesions on the trunk and arms cleared after two-week treatment. c, d Scalp lesions relieved after two-week treatment and cleared after six-week treatment
Discussion
Tinea capitis is a type of fungal infection on the scalp, which primarily affects children aged 3–7 years old [3]. Tinea capitis is uncommon in adults, due to the pH changes and fatty-acids increase in the adult scalp [4]. The proportion of adults among tinea capitis patients was reported to be 2.9% and 4.2% in multi-centre studies from Mexico [5] and Egypt [6], respectively. In single-centre studies, this proportion varies significantly in different countries, from 1.5 to 44.3% [7–11]. In mainland China, adults took up 6.0–13.6% of the tinea capitis population in the 1980s–1990s [12]. This number remained to be 9.0% in the 21th century [13]. In contrast, up to 63% of tinea capitis patients were reported to be adults in a single centre of Taiwan, China [14]. These indicate that adult tinea capitis is becoming less uncommon in some regions, especially in postmenstrual elder women, due to their reduced secretion of fungistatic sebum after menopause. In adult patients, the female-to-male ratio was reported to be 2.2–5.4:1 [4–6, 11, 13, 15] and 26.7–93.5% of the female adults were postmenopausal [10, 11]. Apart from severe immunosuppressive diseases such as HIV infection and post-transplantation, systemic chronic disease including HCV infection (34.4%) and diabetes mellitus (22.4%) has been reported to associate with tinea capitis [6]. Systematic and topical use of corticosteroids are also risk factors [11, 16]. Close contact with animals was reported in 17.1–19.3% of the patients [6, 11, 13]. Furthermore, 20–72% [9, 10, 13] of the adult patients were reported to be immunocompetent. Due to the rise in the occurrence of tinea capitis in immunocompetent adults, adult tinea capitis needs the awareness of clinicians.
Causative agents of adult tinea capitis also vary across geological regions. Trichophyton violaceum was reported to be the main dermatophyte in adults in most countries including Egypt (56.9%, 33/58) [6], Iran (32.0%, 8/25) [10], southern Spain [9] and southern Taiwan, China (74%) [14]. Though M. canis is a major agent causing tinea capitis in children, it is less common in adults [17]. It was reported to account for 15.5% (9/58) infection in Egypt [6] and 16.0% (4/25) in Iran [10]. Exceptionally, M. canis caused 56.5% (13/23) of the adult cases in a single centre in Korea [11]. In mainland China since 2000, Trichophyton violaceum (35.2%, 70/199) was the most common agents, followed by M. canis (21.1%, 42/199), Trichophyton mentagrophyte (16.1%, 32/199) and Trichophyton rubrum (11.5%, 23/199) [13]. Other common dermatophytes isolated in adult patients included Trichophyton tonsurans [5], Trichophyton verrucosum and M. gypseum [6].
Fungal coinfection occurs in 60% of the tinea capitis patients [10]. However, tinea capitis caused by M. canis coexisting with extensive tinea corporis in adults is rare. In this case, the patient had well-controlled diabetes mellitus and denied any other immunosuppression conditions. We hypothesized that M. canis transmitted from animals had stronger pathogenicity which made the skin lesion so generalized. The delay in diagnosis and application of topical corticosteroids also contributed to the lesion generalization. However, the transmission ability of M. canis is weak without animal reservoirs, thus can hardly spread widely among the human population. Its virulence can be lost after about four human-to-human transmissions [3]. This characteristic can be reflected in this case, as the patient did not report any affected family members.
Adult tinea capitis mainly presents as seborrhoeic dermatitis with scaling, grey patch and kerion celsi [11] or pseudo-alopecic plaques [5]. Due to its rarity and atypical presentation, 73.2% of the patients were misdiagnosed during their first consultation in one Korean hospital [11]. Seborrhoeic dermatitis is the diagnosis most commonly confused with tinea capitis in adults [11, 13]. Folliculitis is another major differential diagnosis [4, 11, 13]. The patient in this case was previously misdiagnosed as psoriasis and seborrhoeic dermatitis as these two diseases often present with similar erythema involving scalp and other body areas simultaneously. Thus, a detailed history taking including animal contact and a careful physical examination including searching for broken hair are required. Examinations including Wood’s lamp and dermoscopy can also help in diagnosis and monitoring the treatment effect [18]. Common dermoscopic findings may involve broken hairs, scales, black dots, perifollicular erythema, comma hairs, empty follicles and pustules [11]. In ectothrix infection caused by M.canis, Morse code-like hairs (AKA barcode-like hairs) [19, 20] and white sheaths [21, 22] are typical characteristics under the dermoscope. For a definitive diagnosis, mycological examinations are very important. Among them, 10% KOH microscopic examination of fungal elements is basic for diagnosis and treatment assessment of superficial fungal infection, and novel fluorescent staining can improve the detection rate [23]. Fungal culture can direct the antifungal therapy choice by identification of the dermatophyte.
Treatment of tinea capitis in adults is similar compared with children, yet the age and comorbidities of the patient need to be considered [18]. Terbinafine, griseofulvin and itraconazole have been widely used in the treatment of tinea capitis [10]. As terbinafine hardly influences cytochrome P450 and CYP3A4, it has fewer drug interactions and is increasingly used especially in older patients with multimorbidity [24]. In a meta-analysis comparing the efficacy of griseofulvin and terbinafine in tinea capitis, terbinafine was found to be more effective than griseofulvin for infection by Trichophyton sp., while griseofulvin was shown to be more effective than terbinafine for Microsporum sp. [25]. As M. canis can be resistant to the usual dose of terbinafine (250 mg daily), treatment longer than 4 weeks may be needed for a successful clinical and mycological response [26]. A topical antifungal agent is also advisable for 2-3 months [18]. In this case caused by M.canis, prolonged treatment with oral terbinafine was adopted considering the age and comorbidities of the patient, and the coexisting refractory onychomycosis.
Tinea capitis is not common in adults in most areas. Adult tinea capitis caused by M.canis and co-occurrence with extensive dermatophyte infection are both rarer. This case highlights the challenge in the diagnosis of tinea capitis with extensive dermatophyte infection in adults due to its rarity. In cases presenting as extensive erythema with scales on both scalp and body skin, the fungal infection should be considered. A thorough history including animal contact and physical examination searching for broken hairs are required. Dermoscope, Wood’s lamp examination and further mycological examinations are needed for diagnosis. Terbinafine can be considered as systemic treatment in elderly patients with comorbidities.
Compliance with Ethical Standards
This case report was written complying with the checklist of essential elements instituted by Mycopathologia to guarantee the quality of published case reports [27].
Authors’ Contributions
All the authors involved in the care of the patient. ZY followed up the patient and wrote the text. WC conducted the microscopic examination. ZW and YS involved in the identification of etiologic agent. RL treated the patient and helped to edit the manuscript.
Compliance with Ethical Standards
Conflicts of interest
The authors declare that they have no conflict of interest.
Consent to Participate
Consent from the patient has been obtained and the patient consent form is available.
Consent for Publication
All the authors agree to publish and the written consent for publication has been obtained.
Handling Editor: Vishnu Chaturvedi.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Recovering | ReactionOutcome | CC BY | 33496917 | 20,204,380 | 2021-05 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Drug ineffective'. | Case Report: Massive Spontaneous Pneumothorax-A Rare Form of Presentation for Severe COVID-19 Pneumonia.
Background and Objectives: Coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection is a viral disease that is spreading worldwide and became a pandemic. Although most of the time, the symptoms of the infection are flu like, a percentage of patients develop severe forms, along with severe complications. Many of them are known among front-line health workers, but the number of uncommon presentations and complications has increased. This case report aims to alert healthcare workers on less common forms of presentation, and to introduce this differential diagnosis in the evaluation of patients with COVID-19, given the increasing occurrence of pneumothorax in patients who are not mechanical ventilated. Case presentation: A 57-year-old female patient came to the Emergency Department (ED) by ambulance, with acute respiratory failure. She had SpO2 = 43% on room air at home, and 86% on admission in ED after oxygen delivery (on a reservoir mask). SARS-CoV-2 infection was suspected based on symptoms that started three days ago (fever, dry cough, dyspnea, and fatigability). Blood was taken for lab tests, pharyngeal and nasal swabs for the reverse transcription-PCR (RT-PCR) test, and native computed tomography (CT) was scheduled. The thoracic CT scan showed massive right pneumothorax, partially collapsed lung, multiple bilateral lung infiltrates with a ground glass aspect and the RT-PCR test came back positive for SARS-CoV-2 infection. Despite the prompt diagnosis and treatment of pneumothorax (thoracostomy was performed and the drain tube was placed), the patient died after a long hospitalization in the intensive care unit. Conclusion: Secondary spontaneous pneumothorax (SSP), as a complication in severe forms of COVID-19 pneumonia, especially in female patients without risk factors is rare, and early diagnosis and treatment are essential for increasing the survival chances of these patients.
1. Introduction
Coronavirus disease 2019 (COVID-19) was first reported in Wuhan, Hubei Province, China, in December 2019 [1]. Since its first description, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has led to a pandemic, which was officially declared a global health emergency by the World Health Organization (WHO) on March 11, 2020 [2]. Recent studies have shown that the average age of infected people is around 50 years, even though all ages of the population are susceptible to SARS-CoV-2 infection. However, the clinical manifestations differ according to age [3,4]. A meta-analysis conducted by Peckham et al. which included more than 3 million cases, showed that, although no significant differences were observed between men and women with confirmed COVID-19, male patients were almost three times more likely to be admitted in the intensive care unit due to complications of the disease and a higher risk of death compared to women [5]. The symptoms of SARS-CoV-2 infection have been widely characterized in large studies, with fever, cough, and dyspnea being the most frequent [4,6]. A percentage of patients develop severe forms, along with severe complications.
Spontaneous pneumothorax (SP) is a rare complication of COVID-19 pneumonia, with an incidence of less than 1‰ according to the current literature [7], suggesting that pneumothorax is either uncommon or underreported in patients with COVID-19 [8]. Several case reports have shown that SP associated with COVID-19 pneumonia led to a higher severity and fatal outcome [9]. The prognosis of these patients might be related to the severity of lung lesions, although there is no clear correlation with the magnitude of SP [10]. The timing of SP in the evolution of COVID-19 pneumonia is uncertain. In a recent review of the current literature conducted by Dennison et al. on 32 case reports of spontaneous pneumothorax or pneumomediastinum out of 58 COVID-19 pneumonia infections, SP occurred days to weeks after the onset of symptoms [11]. Therefore, issues like risk factors, timing, outcome, still need to be thoroughly researched.
Here, we describe a case of a 57-year-old woman, who had no prior lung injury or other risk factors for spontaneous pneumothorax, was never a smoker, and was not mechanically ventilated, who developed massive spontaneous pneumothorax after only 3 days of symptoms of SARS-CoV-2 infection.
2. Case Presentation
A 57-years-old woman with a history of essential hypertension was brought into the emergency department (ED) with acute respiratory failure by an ambulance with a doctor. At home, she was found with a significantly decreased peripheral O2 saturation (SpO2) of 43% while breathing room air, saturation that increased to 86% after oxygen delivery (on a reservoir mask). On physical examination the patient presented tachypnea (34 breaths per minute), pale sweaty skin, hemodynamically stable, blood pressure 127/66 mmHg, a heart rate of 109 beats/min, a body temperature of 37.8 °C, Glasgow Coma Score (GCS) = 15/15. The patient complained of fever, dry cough, dyspnea on exertion and fatigue, that started 3 days before the presentation, claiming that, during the previous night, the dyspnea suddenly worsened and an anterior chest pain appeared. She had no other comorbidities except essential hypertension under treatment. Based on the history and clinical examination, SARS-CoV-2 infection was suspected. Prone position was initiated in order to improve ventilation, which the patient did not tolerate, so she returned to the seated position. She received an intravenous perfusion with 1 g acetaminophen, 8 mg dexamethasone, 1500 mg vitamin C, 40 mg pantoprazole and 500 mL normal saline (0.9%) solution. Blood was taken for lab tests and she was scheduled for a thoracic CT (computed tomography) scan. The pharyngeal and nasal swabs were taken for the RT-PCR test, which came positive after several hours.
The thoracic CT scan showed massive right pneumothorax (90 mm), lung partially collapsed, slightly left-displaced heart, multiple bilateral lung infiltrates with a ground-glass aspect that occupied about 65% of lung fields—CO-RADS classification 5, typical COVID-19 aspect. (Figure 1)
The thoracostomy was performed and the drain tube was placed in the right fifth intercostal space, the medium axillary line, under local anesthesia. The patient’s condition partially improved and oxygen saturation increased by approximately 6% (SpO2 = 92–93%) 15–20 min after the drain tube was placed. Oxygen therapy on a reservoir mask with 15 L/min was continued, and remdesivir 200 mg as loading dose was initiated, then 100 mg at 24 h, lopinavir/ritonavir 200 mg/50 mg—2 pills every 12 h, enoxaparin 60 mg twice daily, dexamethasone 8 mg three times daily, pantoprazole 40 mg twice daily, vitamin C 1 g every 6 h, Ceftriaxone 1 g twice daily, her previous medicines for hypertension, acetaminophen on need and soluble regular insulin according to her glycemia level (given the constant high values observed during her hospitalization), vitamin D, B1, B6, zinc and alprazolam.
Laboratory test results found to be pathological are shown in Table 1.
After a few hours, a control chest X-ray was performed, confirming the correct placement of the drain tube with full expansion of the collapsed lung (Figure 2). During the day, the patient’s breathing worsened (tachypnea 40 breaths/minute) with increased breathing effort, decreased oxygen saturation to 70%, hypoxemia (pO2 = 36 mmHg on arterial blood gases (ABG)) despite maximum oxygen delivery so she was admitted to the intensive care unit and non-invasive ventilation (NIV) was initiated. The patient tolerated non-invasive ventilation for a few hours, but then the condition of the patient worsened again, and the endotracheal intubation and invasive mechanical ventilation were decided.
During hospitalization in the intensive care unit, the patient had a fluctuating evolution, and despite the treatment (medication, kinetotherapy, prone position, hydro-electrolyte rebalance solutions), the patient’s condition was deteriorating and required vasopressor support and mechanical ventilation. At 19 days after admission in the intensive care unit, the patient had cardiac arrest through asystole and did not respond to resuscitation maneuvers.
3. Discussion
Spontaneous pneumothorax is a type of pneumothorax that develops in the absence of trauma [12]. It is classified as primary and secondary SP. Primary spontaneous pneumothorax (PSP) occurs in patients without pre-existing lung disease, compared with secondary spontaneous pneumothorax (SSP), which is a complication occurring in an affected lung. PSP is common in young adults, with a higher incidence in men than in women (7.4 to 18 per 100,000 men and 1.2 to 6 per 100,000 women) [13].
Risk factors for PSP include male gender, tall and thin stature and smoking [14]. In SSP, the most common underlying disorders are COPD with a predominance of pulmonary emphysema, cystic fibrosis, tuberculosis, lung cancer, interstitial pneumonitis, and Pneumocystis carinii pneumonia associated with human immunodeficiency virus. While PSP usually occurs between the ages of 10 and 30, the maximum incidence of SSP is observed in the following years—between the ages of 60 and 64—depending on the baseline condition [12]. Identifying the cause of an SSP is crucial, as the immediate and long-term management of the SSP differs from that of the PSP, along with significantly more serious consequences [15].
Massive pneumothorax is a major, life-threatening emergency that must be identified and treated very promptly [16], regardless if the patient presents to ED or it occurs as a complication in COVID 19 ward patients. Our patient had a massive pneumothorax that was timely solved, but still she had a fatal outcome. However, we cannot state that the size of the pneumothorax had a defining role in the patient’s evolution as there are studies with COVID-19 patients experiencing small pneumothorax and still a poor prognosis [9,10].
Further research is needed in order to identify other possible factors influencing the evolution of patients with COVID-19 pneumonia and SP. In a large study by Òscar Miró et al., the most frequent met comorbidities in patients with COVID-19 and SP were hypertension (37.5% of cases), asthma (20% of cases) and diabetes (17.5% of cases) [7]. Our patient had both hypertension and diabetes (not previously known); however, we cannot state a direct relation between these factors and SP development. Nevertheless, these comorbidities and the lesions’ extension were determinant for the patient’s outcome.
Spontaneous pneumothorax is a commonly known complication in patients with acute respiratory distress syndrome (ARDS), where the most frequent causes are pressure and volume-related alveolar rupture [17]. Histological examination of lung biopsy samples in a patient who died from COVID-19 pneumonia showed desquamation of pneumocytes and hyaline membrane formation, indicating ARDS [15].
We suspect that our patient developed SSP because of lung lesions caused by COVID-19 infection, given the fact that she was not a smoker and not known to have other risk factors. The patient had sudden onset dyspnea, chest pain, tachycardia, hypoxia, and increased D-Dimers, so one of the considered differential diagnosis was pulmonary embolism. However, it was initially decided to perform a native thoracic CT scan, which led to the correct diagnosis and completely changed the patient’s therapeutic management, being the first patient with such a complication of SARS-CoV-2 infection in our clinic. As opposed to different case reports from the literature, where the CT scans were performed late in the course of patients’ evolution and dictated by the deterioration of the patient, in our case, the CT scan was performed from the admittance to the ED [9,11]. The large availability of CT scan in our hospital helped us to better quantify the overall severity of patients with COVID-19 pneumonia and led to a prompt management of our case.
Despite the timely diagnosis and treatment of SSP, the patient’s evolution was marked by severe lung damage from COVID-19 pneumonia, and after a long hospitalization in the intensive care unit, the patient died, emphasizing the importance of the underlying lung disease in spontaneous secondary pneumothorax. Patients with COVID-19 pneumonia and SP should be carefully monitored to prevent respiratory deterioration, no matter the size of the pneumothorax [10].
4. Conclusions
Secondary spontaneous pneumothorax should always be considered as a differential diagnosis in the assessment of patients with SARS-CoV-2 infection and acute respiratory failure. Prompt diagnosis and treatment are crucial in the further evolution of the patient. Large studies are needed to measure the prognosis of these pulmonary complications in patients with SARS-CoV-2 infection, but since the cases are sporadic, any detailed report may have an added value.
Author Contributions
Conceptualization, A.M.M.; writing—original draft preparation, A.M.M.; writing—review and editing, A.P., F.N.B., A.M.M.; validation, O.A.M.; visualization, A.P., O.A.M.; supervision, O.A.M. All authors have read and agreed to the published version of the manuscript.
Funding
This research received no external funding.
Institutional Review Board Statement
The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Institutional Review Board of Emergency Clinical Municipal Hospital, 300041 Timisoara, Romania (number E-5947/29 December 2020).
Informed Consent Statement
Written informed consent has been obtained from the patient to publish this paper.
Data Availability Statement
The data presented in this study are available on request from the corresponding author. The data are not publicly available due to patient confidentiality.
Conflicts of Interest
The authors declare no conflict of interest.
Figure 1 CT scan at admission in ED: massive right pneumothorax, partially collapsed lung, slightly left-displaced heart, multiple bilateral ground glass lung infiltrates.
Figure 2 Chest radiography of the patient the day after insertion of the drain tube, demonstrating the re-expansion of the lung and multiple bilateral opacities in the context of COVID-19 infection.
medicina-57-00082-t001_Table 1 Table 1 Pathological values of the laboratory tests performed in the Emergency Department.
Laboratory Test Conventional Units Value Reference Range Value
WBC ×109/µL 3.8 4–10
Lymphocytes ×103/µL 0.7 1–3
AST U/L 107 2–32
ALT U/L 80 2–33
Ferritin ng/mL 1463 15–150
Blood glucose mg/dL 361 74–106
CRP mg/dL 85.9 0–5
D-dimers ng/mL 1212.8 age × 10
aPTT seconds extremely low * 25–36
WBC = white blood cells; AST = aspartate aminotransferase; ALT = alanine aminotransferase; CRP = C-reactive protein; aPTT = activated partial thromboplastin time. * value under the detection limit of the equipment.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. | ACETAMINOPHEN, ASCORBIC ACID, DEXAMETHASONE, OXYGEN, PANTOPRAZOLE, REMDESIVIR, SODIUM CHLORIDE | DrugsGivenReaction | CC BY | 33498180 | 18,963,873 | 2021-01-20 |
What was the administration route of drug 'REMDESIVIR'? | Case Report: Massive Spontaneous Pneumothorax-A Rare Form of Presentation for Severe COVID-19 Pneumonia.
Background and Objectives: Coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection is a viral disease that is spreading worldwide and became a pandemic. Although most of the time, the symptoms of the infection are flu like, a percentage of patients develop severe forms, along with severe complications. Many of them are known among front-line health workers, but the number of uncommon presentations and complications has increased. This case report aims to alert healthcare workers on less common forms of presentation, and to introduce this differential diagnosis in the evaluation of patients with COVID-19, given the increasing occurrence of pneumothorax in patients who are not mechanical ventilated. Case presentation: A 57-year-old female patient came to the Emergency Department (ED) by ambulance, with acute respiratory failure. She had SpO2 = 43% on room air at home, and 86% on admission in ED after oxygen delivery (on a reservoir mask). SARS-CoV-2 infection was suspected based on symptoms that started three days ago (fever, dry cough, dyspnea, and fatigability). Blood was taken for lab tests, pharyngeal and nasal swabs for the reverse transcription-PCR (RT-PCR) test, and native computed tomography (CT) was scheduled. The thoracic CT scan showed massive right pneumothorax, partially collapsed lung, multiple bilateral lung infiltrates with a ground glass aspect and the RT-PCR test came back positive for SARS-CoV-2 infection. Despite the prompt diagnosis and treatment of pneumothorax (thoracostomy was performed and the drain tube was placed), the patient died after a long hospitalization in the intensive care unit. Conclusion: Secondary spontaneous pneumothorax (SSP), as a complication in severe forms of COVID-19 pneumonia, especially in female patients without risk factors is rare, and early diagnosis and treatment are essential for increasing the survival chances of these patients.
1. Introduction
Coronavirus disease 2019 (COVID-19) was first reported in Wuhan, Hubei Province, China, in December 2019 [1]. Since its first description, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has led to a pandemic, which was officially declared a global health emergency by the World Health Organization (WHO) on March 11, 2020 [2]. Recent studies have shown that the average age of infected people is around 50 years, even though all ages of the population are susceptible to SARS-CoV-2 infection. However, the clinical manifestations differ according to age [3,4]. A meta-analysis conducted by Peckham et al. which included more than 3 million cases, showed that, although no significant differences were observed between men and women with confirmed COVID-19, male patients were almost three times more likely to be admitted in the intensive care unit due to complications of the disease and a higher risk of death compared to women [5]. The symptoms of SARS-CoV-2 infection have been widely characterized in large studies, with fever, cough, and dyspnea being the most frequent [4,6]. A percentage of patients develop severe forms, along with severe complications.
Spontaneous pneumothorax (SP) is a rare complication of COVID-19 pneumonia, with an incidence of less than 1‰ according to the current literature [7], suggesting that pneumothorax is either uncommon or underreported in patients with COVID-19 [8]. Several case reports have shown that SP associated with COVID-19 pneumonia led to a higher severity and fatal outcome [9]. The prognosis of these patients might be related to the severity of lung lesions, although there is no clear correlation with the magnitude of SP [10]. The timing of SP in the evolution of COVID-19 pneumonia is uncertain. In a recent review of the current literature conducted by Dennison et al. on 32 case reports of spontaneous pneumothorax or pneumomediastinum out of 58 COVID-19 pneumonia infections, SP occurred days to weeks after the onset of symptoms [11]. Therefore, issues like risk factors, timing, outcome, still need to be thoroughly researched.
Here, we describe a case of a 57-year-old woman, who had no prior lung injury or other risk factors for spontaneous pneumothorax, was never a smoker, and was not mechanically ventilated, who developed massive spontaneous pneumothorax after only 3 days of symptoms of SARS-CoV-2 infection.
2. Case Presentation
A 57-years-old woman with a history of essential hypertension was brought into the emergency department (ED) with acute respiratory failure by an ambulance with a doctor. At home, she was found with a significantly decreased peripheral O2 saturation (SpO2) of 43% while breathing room air, saturation that increased to 86% after oxygen delivery (on a reservoir mask). On physical examination the patient presented tachypnea (34 breaths per minute), pale sweaty skin, hemodynamically stable, blood pressure 127/66 mmHg, a heart rate of 109 beats/min, a body temperature of 37.8 °C, Glasgow Coma Score (GCS) = 15/15. The patient complained of fever, dry cough, dyspnea on exertion and fatigue, that started 3 days before the presentation, claiming that, during the previous night, the dyspnea suddenly worsened and an anterior chest pain appeared. She had no other comorbidities except essential hypertension under treatment. Based on the history and clinical examination, SARS-CoV-2 infection was suspected. Prone position was initiated in order to improve ventilation, which the patient did not tolerate, so she returned to the seated position. She received an intravenous perfusion with 1 g acetaminophen, 8 mg dexamethasone, 1500 mg vitamin C, 40 mg pantoprazole and 500 mL normal saline (0.9%) solution. Blood was taken for lab tests and she was scheduled for a thoracic CT (computed tomography) scan. The pharyngeal and nasal swabs were taken for the RT-PCR test, which came positive after several hours.
The thoracic CT scan showed massive right pneumothorax (90 mm), lung partially collapsed, slightly left-displaced heart, multiple bilateral lung infiltrates with a ground-glass aspect that occupied about 65% of lung fields—CO-RADS classification 5, typical COVID-19 aspect. (Figure 1)
The thoracostomy was performed and the drain tube was placed in the right fifth intercostal space, the medium axillary line, under local anesthesia. The patient’s condition partially improved and oxygen saturation increased by approximately 6% (SpO2 = 92–93%) 15–20 min after the drain tube was placed. Oxygen therapy on a reservoir mask with 15 L/min was continued, and remdesivir 200 mg as loading dose was initiated, then 100 mg at 24 h, lopinavir/ritonavir 200 mg/50 mg—2 pills every 12 h, enoxaparin 60 mg twice daily, dexamethasone 8 mg three times daily, pantoprazole 40 mg twice daily, vitamin C 1 g every 6 h, Ceftriaxone 1 g twice daily, her previous medicines for hypertension, acetaminophen on need and soluble regular insulin according to her glycemia level (given the constant high values observed during her hospitalization), vitamin D, B1, B6, zinc and alprazolam.
Laboratory test results found to be pathological are shown in Table 1.
After a few hours, a control chest X-ray was performed, confirming the correct placement of the drain tube with full expansion of the collapsed lung (Figure 2). During the day, the patient’s breathing worsened (tachypnea 40 breaths/minute) with increased breathing effort, decreased oxygen saturation to 70%, hypoxemia (pO2 = 36 mmHg on arterial blood gases (ABG)) despite maximum oxygen delivery so she was admitted to the intensive care unit and non-invasive ventilation (NIV) was initiated. The patient tolerated non-invasive ventilation for a few hours, but then the condition of the patient worsened again, and the endotracheal intubation and invasive mechanical ventilation were decided.
During hospitalization in the intensive care unit, the patient had a fluctuating evolution, and despite the treatment (medication, kinetotherapy, prone position, hydro-electrolyte rebalance solutions), the patient’s condition was deteriorating and required vasopressor support and mechanical ventilation. At 19 days after admission in the intensive care unit, the patient had cardiac arrest through asystole and did not respond to resuscitation maneuvers.
3. Discussion
Spontaneous pneumothorax is a type of pneumothorax that develops in the absence of trauma [12]. It is classified as primary and secondary SP. Primary spontaneous pneumothorax (PSP) occurs in patients without pre-existing lung disease, compared with secondary spontaneous pneumothorax (SSP), which is a complication occurring in an affected lung. PSP is common in young adults, with a higher incidence in men than in women (7.4 to 18 per 100,000 men and 1.2 to 6 per 100,000 women) [13].
Risk factors for PSP include male gender, tall and thin stature and smoking [14]. In SSP, the most common underlying disorders are COPD with a predominance of pulmonary emphysema, cystic fibrosis, tuberculosis, lung cancer, interstitial pneumonitis, and Pneumocystis carinii pneumonia associated with human immunodeficiency virus. While PSP usually occurs between the ages of 10 and 30, the maximum incidence of SSP is observed in the following years—between the ages of 60 and 64—depending on the baseline condition [12]. Identifying the cause of an SSP is crucial, as the immediate and long-term management of the SSP differs from that of the PSP, along with significantly more serious consequences [15].
Massive pneumothorax is a major, life-threatening emergency that must be identified and treated very promptly [16], regardless if the patient presents to ED or it occurs as a complication in COVID 19 ward patients. Our patient had a massive pneumothorax that was timely solved, but still she had a fatal outcome. However, we cannot state that the size of the pneumothorax had a defining role in the patient’s evolution as there are studies with COVID-19 patients experiencing small pneumothorax and still a poor prognosis [9,10].
Further research is needed in order to identify other possible factors influencing the evolution of patients with COVID-19 pneumonia and SP. In a large study by Òscar Miró et al., the most frequent met comorbidities in patients with COVID-19 and SP were hypertension (37.5% of cases), asthma (20% of cases) and diabetes (17.5% of cases) [7]. Our patient had both hypertension and diabetes (not previously known); however, we cannot state a direct relation between these factors and SP development. Nevertheless, these comorbidities and the lesions’ extension were determinant for the patient’s outcome.
Spontaneous pneumothorax is a commonly known complication in patients with acute respiratory distress syndrome (ARDS), where the most frequent causes are pressure and volume-related alveolar rupture [17]. Histological examination of lung biopsy samples in a patient who died from COVID-19 pneumonia showed desquamation of pneumocytes and hyaline membrane formation, indicating ARDS [15].
We suspect that our patient developed SSP because of lung lesions caused by COVID-19 infection, given the fact that she was not a smoker and not known to have other risk factors. The patient had sudden onset dyspnea, chest pain, tachycardia, hypoxia, and increased D-Dimers, so one of the considered differential diagnosis was pulmonary embolism. However, it was initially decided to perform a native thoracic CT scan, which led to the correct diagnosis and completely changed the patient’s therapeutic management, being the first patient with such a complication of SARS-CoV-2 infection in our clinic. As opposed to different case reports from the literature, where the CT scans were performed late in the course of patients’ evolution and dictated by the deterioration of the patient, in our case, the CT scan was performed from the admittance to the ED [9,11]. The large availability of CT scan in our hospital helped us to better quantify the overall severity of patients with COVID-19 pneumonia and led to a prompt management of our case.
Despite the timely diagnosis and treatment of SSP, the patient’s evolution was marked by severe lung damage from COVID-19 pneumonia, and after a long hospitalization in the intensive care unit, the patient died, emphasizing the importance of the underlying lung disease in spontaneous secondary pneumothorax. Patients with COVID-19 pneumonia and SP should be carefully monitored to prevent respiratory deterioration, no matter the size of the pneumothorax [10].
4. Conclusions
Secondary spontaneous pneumothorax should always be considered as a differential diagnosis in the assessment of patients with SARS-CoV-2 infection and acute respiratory failure. Prompt diagnosis and treatment are crucial in the further evolution of the patient. Large studies are needed to measure the prognosis of these pulmonary complications in patients with SARS-CoV-2 infection, but since the cases are sporadic, any detailed report may have an added value.
Author Contributions
Conceptualization, A.M.M.; writing—original draft preparation, A.M.M.; writing—review and editing, A.P., F.N.B., A.M.M.; validation, O.A.M.; visualization, A.P., O.A.M.; supervision, O.A.M. All authors have read and agreed to the published version of the manuscript.
Funding
This research received no external funding.
Institutional Review Board Statement
The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Institutional Review Board of Emergency Clinical Municipal Hospital, 300041 Timisoara, Romania (number E-5947/29 December 2020).
Informed Consent Statement
Written informed consent has been obtained from the patient to publish this paper.
Data Availability Statement
The data presented in this study are available on request from the corresponding author. The data are not publicly available due to patient confidentiality.
Conflicts of Interest
The authors declare no conflict of interest.
Figure 1 CT scan at admission in ED: massive right pneumothorax, partially collapsed lung, slightly left-displaced heart, multiple bilateral ground glass lung infiltrates.
Figure 2 Chest radiography of the patient the day after insertion of the drain tube, demonstrating the re-expansion of the lung and multiple bilateral opacities in the context of COVID-19 infection.
medicina-57-00082-t001_Table 1 Table 1 Pathological values of the laboratory tests performed in the Emergency Department.
Laboratory Test Conventional Units Value Reference Range Value
WBC ×109/µL 3.8 4–10
Lymphocytes ×103/µL 0.7 1–3
AST U/L 107 2–32
ALT U/L 80 2–33
Ferritin ng/mL 1463 15–150
Blood glucose mg/dL 361 74–106
CRP mg/dL 85.9 0–5
D-dimers ng/mL 1212.8 age × 10
aPTT seconds extremely low * 25–36
WBC = white blood cells; AST = aspartate aminotransferase; ALT = alanine aminotransferase; CRP = C-reactive protein; aPTT = activated partial thromboplastin time. * value under the detection limit of the equipment.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Intravenous (not otherwise specified) | DrugAdministrationRoute | CC BY | 33498180 | 18,963,873 | 2021-01-20 |
What was the outcome of reaction 'Drug ineffective'? | Case Report: Massive Spontaneous Pneumothorax-A Rare Form of Presentation for Severe COVID-19 Pneumonia.
Background and Objectives: Coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection is a viral disease that is spreading worldwide and became a pandemic. Although most of the time, the symptoms of the infection are flu like, a percentage of patients develop severe forms, along with severe complications. Many of them are known among front-line health workers, but the number of uncommon presentations and complications has increased. This case report aims to alert healthcare workers on less common forms of presentation, and to introduce this differential diagnosis in the evaluation of patients with COVID-19, given the increasing occurrence of pneumothorax in patients who are not mechanical ventilated. Case presentation: A 57-year-old female patient came to the Emergency Department (ED) by ambulance, with acute respiratory failure. She had SpO2 = 43% on room air at home, and 86% on admission in ED after oxygen delivery (on a reservoir mask). SARS-CoV-2 infection was suspected based on symptoms that started three days ago (fever, dry cough, dyspnea, and fatigability). Blood was taken for lab tests, pharyngeal and nasal swabs for the reverse transcription-PCR (RT-PCR) test, and native computed tomography (CT) was scheduled. The thoracic CT scan showed massive right pneumothorax, partially collapsed lung, multiple bilateral lung infiltrates with a ground glass aspect and the RT-PCR test came back positive for SARS-CoV-2 infection. Despite the prompt diagnosis and treatment of pneumothorax (thoracostomy was performed and the drain tube was placed), the patient died after a long hospitalization in the intensive care unit. Conclusion: Secondary spontaneous pneumothorax (SSP), as a complication in severe forms of COVID-19 pneumonia, especially in female patients without risk factors is rare, and early diagnosis and treatment are essential for increasing the survival chances of these patients.
1. Introduction
Coronavirus disease 2019 (COVID-19) was first reported in Wuhan, Hubei Province, China, in December 2019 [1]. Since its first description, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has led to a pandemic, which was officially declared a global health emergency by the World Health Organization (WHO) on March 11, 2020 [2]. Recent studies have shown that the average age of infected people is around 50 years, even though all ages of the population are susceptible to SARS-CoV-2 infection. However, the clinical manifestations differ according to age [3,4]. A meta-analysis conducted by Peckham et al. which included more than 3 million cases, showed that, although no significant differences were observed between men and women with confirmed COVID-19, male patients were almost three times more likely to be admitted in the intensive care unit due to complications of the disease and a higher risk of death compared to women [5]. The symptoms of SARS-CoV-2 infection have been widely characterized in large studies, with fever, cough, and dyspnea being the most frequent [4,6]. A percentage of patients develop severe forms, along with severe complications.
Spontaneous pneumothorax (SP) is a rare complication of COVID-19 pneumonia, with an incidence of less than 1‰ according to the current literature [7], suggesting that pneumothorax is either uncommon or underreported in patients with COVID-19 [8]. Several case reports have shown that SP associated with COVID-19 pneumonia led to a higher severity and fatal outcome [9]. The prognosis of these patients might be related to the severity of lung lesions, although there is no clear correlation with the magnitude of SP [10]. The timing of SP in the evolution of COVID-19 pneumonia is uncertain. In a recent review of the current literature conducted by Dennison et al. on 32 case reports of spontaneous pneumothorax or pneumomediastinum out of 58 COVID-19 pneumonia infections, SP occurred days to weeks after the onset of symptoms [11]. Therefore, issues like risk factors, timing, outcome, still need to be thoroughly researched.
Here, we describe a case of a 57-year-old woman, who had no prior lung injury or other risk factors for spontaneous pneumothorax, was never a smoker, and was not mechanically ventilated, who developed massive spontaneous pneumothorax after only 3 days of symptoms of SARS-CoV-2 infection.
2. Case Presentation
A 57-years-old woman with a history of essential hypertension was brought into the emergency department (ED) with acute respiratory failure by an ambulance with a doctor. At home, she was found with a significantly decreased peripheral O2 saturation (SpO2) of 43% while breathing room air, saturation that increased to 86% after oxygen delivery (on a reservoir mask). On physical examination the patient presented tachypnea (34 breaths per minute), pale sweaty skin, hemodynamically stable, blood pressure 127/66 mmHg, a heart rate of 109 beats/min, a body temperature of 37.8 °C, Glasgow Coma Score (GCS) = 15/15. The patient complained of fever, dry cough, dyspnea on exertion and fatigue, that started 3 days before the presentation, claiming that, during the previous night, the dyspnea suddenly worsened and an anterior chest pain appeared. She had no other comorbidities except essential hypertension under treatment. Based on the history and clinical examination, SARS-CoV-2 infection was suspected. Prone position was initiated in order to improve ventilation, which the patient did not tolerate, so she returned to the seated position. She received an intravenous perfusion with 1 g acetaminophen, 8 mg dexamethasone, 1500 mg vitamin C, 40 mg pantoprazole and 500 mL normal saline (0.9%) solution. Blood was taken for lab tests and she was scheduled for a thoracic CT (computed tomography) scan. The pharyngeal and nasal swabs were taken for the RT-PCR test, which came positive after several hours.
The thoracic CT scan showed massive right pneumothorax (90 mm), lung partially collapsed, slightly left-displaced heart, multiple bilateral lung infiltrates with a ground-glass aspect that occupied about 65% of lung fields—CO-RADS classification 5, typical COVID-19 aspect. (Figure 1)
The thoracostomy was performed and the drain tube was placed in the right fifth intercostal space, the medium axillary line, under local anesthesia. The patient’s condition partially improved and oxygen saturation increased by approximately 6% (SpO2 = 92–93%) 15–20 min after the drain tube was placed. Oxygen therapy on a reservoir mask with 15 L/min was continued, and remdesivir 200 mg as loading dose was initiated, then 100 mg at 24 h, lopinavir/ritonavir 200 mg/50 mg—2 pills every 12 h, enoxaparin 60 mg twice daily, dexamethasone 8 mg three times daily, pantoprazole 40 mg twice daily, vitamin C 1 g every 6 h, Ceftriaxone 1 g twice daily, her previous medicines for hypertension, acetaminophen on need and soluble regular insulin according to her glycemia level (given the constant high values observed during her hospitalization), vitamin D, B1, B6, zinc and alprazolam.
Laboratory test results found to be pathological are shown in Table 1.
After a few hours, a control chest X-ray was performed, confirming the correct placement of the drain tube with full expansion of the collapsed lung (Figure 2). During the day, the patient’s breathing worsened (tachypnea 40 breaths/minute) with increased breathing effort, decreased oxygen saturation to 70%, hypoxemia (pO2 = 36 mmHg on arterial blood gases (ABG)) despite maximum oxygen delivery so she was admitted to the intensive care unit and non-invasive ventilation (NIV) was initiated. The patient tolerated non-invasive ventilation for a few hours, but then the condition of the patient worsened again, and the endotracheal intubation and invasive mechanical ventilation were decided.
During hospitalization in the intensive care unit, the patient had a fluctuating evolution, and despite the treatment (medication, kinetotherapy, prone position, hydro-electrolyte rebalance solutions), the patient’s condition was deteriorating and required vasopressor support and mechanical ventilation. At 19 days after admission in the intensive care unit, the patient had cardiac arrest through asystole and did not respond to resuscitation maneuvers.
3. Discussion
Spontaneous pneumothorax is a type of pneumothorax that develops in the absence of trauma [12]. It is classified as primary and secondary SP. Primary spontaneous pneumothorax (PSP) occurs in patients without pre-existing lung disease, compared with secondary spontaneous pneumothorax (SSP), which is a complication occurring in an affected lung. PSP is common in young adults, with a higher incidence in men than in women (7.4 to 18 per 100,000 men and 1.2 to 6 per 100,000 women) [13].
Risk factors for PSP include male gender, tall and thin stature and smoking [14]. In SSP, the most common underlying disorders are COPD with a predominance of pulmonary emphysema, cystic fibrosis, tuberculosis, lung cancer, interstitial pneumonitis, and Pneumocystis carinii pneumonia associated with human immunodeficiency virus. While PSP usually occurs between the ages of 10 and 30, the maximum incidence of SSP is observed in the following years—between the ages of 60 and 64—depending on the baseline condition [12]. Identifying the cause of an SSP is crucial, as the immediate and long-term management of the SSP differs from that of the PSP, along with significantly more serious consequences [15].
Massive pneumothorax is a major, life-threatening emergency that must be identified and treated very promptly [16], regardless if the patient presents to ED or it occurs as a complication in COVID 19 ward patients. Our patient had a massive pneumothorax that was timely solved, but still she had a fatal outcome. However, we cannot state that the size of the pneumothorax had a defining role in the patient’s evolution as there are studies with COVID-19 patients experiencing small pneumothorax and still a poor prognosis [9,10].
Further research is needed in order to identify other possible factors influencing the evolution of patients with COVID-19 pneumonia and SP. In a large study by Òscar Miró et al., the most frequent met comorbidities in patients with COVID-19 and SP were hypertension (37.5% of cases), asthma (20% of cases) and diabetes (17.5% of cases) [7]. Our patient had both hypertension and diabetes (not previously known); however, we cannot state a direct relation between these factors and SP development. Nevertheless, these comorbidities and the lesions’ extension were determinant for the patient’s outcome.
Spontaneous pneumothorax is a commonly known complication in patients with acute respiratory distress syndrome (ARDS), where the most frequent causes are pressure and volume-related alveolar rupture [17]. Histological examination of lung biopsy samples in a patient who died from COVID-19 pneumonia showed desquamation of pneumocytes and hyaline membrane formation, indicating ARDS [15].
We suspect that our patient developed SSP because of lung lesions caused by COVID-19 infection, given the fact that she was not a smoker and not known to have other risk factors. The patient had sudden onset dyspnea, chest pain, tachycardia, hypoxia, and increased D-Dimers, so one of the considered differential diagnosis was pulmonary embolism. However, it was initially decided to perform a native thoracic CT scan, which led to the correct diagnosis and completely changed the patient’s therapeutic management, being the first patient with such a complication of SARS-CoV-2 infection in our clinic. As opposed to different case reports from the literature, where the CT scans were performed late in the course of patients’ evolution and dictated by the deterioration of the patient, in our case, the CT scan was performed from the admittance to the ED [9,11]. The large availability of CT scan in our hospital helped us to better quantify the overall severity of patients with COVID-19 pneumonia and led to a prompt management of our case.
Despite the timely diagnosis and treatment of SSP, the patient’s evolution was marked by severe lung damage from COVID-19 pneumonia, and after a long hospitalization in the intensive care unit, the patient died, emphasizing the importance of the underlying lung disease in spontaneous secondary pneumothorax. Patients with COVID-19 pneumonia and SP should be carefully monitored to prevent respiratory deterioration, no matter the size of the pneumothorax [10].
4. Conclusions
Secondary spontaneous pneumothorax should always be considered as a differential diagnosis in the assessment of patients with SARS-CoV-2 infection and acute respiratory failure. Prompt diagnosis and treatment are crucial in the further evolution of the patient. Large studies are needed to measure the prognosis of these pulmonary complications in patients with SARS-CoV-2 infection, but since the cases are sporadic, any detailed report may have an added value.
Author Contributions
Conceptualization, A.M.M.; writing—original draft preparation, A.M.M.; writing—review and editing, A.P., F.N.B., A.M.M.; validation, O.A.M.; visualization, A.P., O.A.M.; supervision, O.A.M. All authors have read and agreed to the published version of the manuscript.
Funding
This research received no external funding.
Institutional Review Board Statement
The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Institutional Review Board of Emergency Clinical Municipal Hospital, 300041 Timisoara, Romania (number E-5947/29 December 2020).
Informed Consent Statement
Written informed consent has been obtained from the patient to publish this paper.
Data Availability Statement
The data presented in this study are available on request from the corresponding author. The data are not publicly available due to patient confidentiality.
Conflicts of Interest
The authors declare no conflict of interest.
Figure 1 CT scan at admission in ED: massive right pneumothorax, partially collapsed lung, slightly left-displaced heart, multiple bilateral ground glass lung infiltrates.
Figure 2 Chest radiography of the patient the day after insertion of the drain tube, demonstrating the re-expansion of the lung and multiple bilateral opacities in the context of COVID-19 infection.
medicina-57-00082-t001_Table 1 Table 1 Pathological values of the laboratory tests performed in the Emergency Department.
Laboratory Test Conventional Units Value Reference Range Value
WBC ×109/µL 3.8 4–10
Lymphocytes ×103/µL 0.7 1–3
AST U/L 107 2–32
ALT U/L 80 2–33
Ferritin ng/mL 1463 15–150
Blood glucose mg/dL 361 74–106
CRP mg/dL 85.9 0–5
D-dimers ng/mL 1212.8 age × 10
aPTT seconds extremely low * 25–36
WBC = white blood cells; AST = aspartate aminotransferase; ALT = alanine aminotransferase; CRP = C-reactive protein; aPTT = activated partial thromboplastin time. * value under the detection limit of the equipment.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Fatal | ReactionOutcome | CC BY | 33498180 | 18,963,873 | 2021-01-20 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'No adverse event'. | Pharmacogenetic Testing of Cytochrome P450 Drug Metabolizing Enzymes in a Case Series of Patients with Prader-Willi Syndrome.
Prader-Willi syndrome (PWS) is associated with co-morbid psychiatric symptoms (disruptive behavior, anxiety, mood disorders, and psychosis) often requiring psychotropic medications. In this clinical case series of 35 patients with PWS, pharmacogenetic testing was obtained to determine allele frequencies predicting variations in activity of cytochrome (CYP) P450 drug metabolizing enzymes 2D6, 2B6, 2C19, 2C9, 3A4, and 1A2. Results were deidentified, collated, and analyzed by PWS genetic subtype: 14 deletion (DEL), 16 maternal uniparental disomy (UPD) and 5 DNA-methylation positive unspecified molecular subtype (PWS Unspec). Literature review informed comparative population frequencies of CYP polymorphisms, phenotypes, and substrate specificity. Among the total PWS cohort, extensive metabolizer (EM) activity prevailed across all cytochromes except CYP1A2, which showed greater ultra-rapid metabolizer (UM) status (p < 0.05), especially among UPD. Among PWS genetic subtypes, there were statistically significant differences in metabolizing status for cytochromes 2D6, 2C19, 2C9, 3A4 and 1A2 acting on substrates such as fluoxetine, risperidone, sertraline, modafinil, aripiprazole, citalopram, and escitalopram. Gonadal steroid therapy may further impact metabolism of 2C19, 2C9, 3A4 and 1A2 substrates. The status of growth hormone treatment may affect CYP3A4 activity with gender specificity. Pharmacogenetic testing together with PWS genetic subtyping may inform psychotropic medication dosing parameters and risk for adverse events.
1. Introduction
1.1. Prader-Willi Syndrome (PWS)
PWS is a rare, complex genetic disorder recognized as the most common form of syndromic obesity. PWS is reported to affect between 350,000 and 400,000 individuals worldwide with an estimated prevalence of one in 10,000 to 38,000 individuals [1]. PWS results from errors in genomic imprinting with loss of expression of paternal genes in the chromosome 15q11–q13 region generally from a paternal deletion (DEL, 60% of cases) followed by maternal uniparental disomy 15 (UPD, 35% of cases) in which both chromosome 15 s are inherited from the mother. Imprinting center defects (IC) and chromosome 15 translocations or inversions are seen in the remaining patients [2]. This multi-system disorder is characterized by severe infantile hypotonia with a poor suck, weak cry, failure to thrive and feeding difficulties. Hypogonadism/hypogenitalism is noted at birth. Hypothalamic dysfunction causes growth and other hormone deficiencies, as well as dysregulation of body temperature, sleep and wakefulness, hunger, thirst, and stress response. Currently, growth hormone, usually started in the first year of life, is the only medication indicated for PWS to manage short stature, small hands and feet, global developmental delay and abnormalities of body composition consisting of excessive fat mass to lean body weight [3]. In early childhood there is an increased interest in food, and calories must be managed to avoid excess weight gain. Six nutritional phases have been described [4]. Excessive food intake leads to obesity, unless dietary management, environmental controls and mandatory exercise are implemented. Psychological food security helps to manage behavior difficulties presenting around food [5]. Temper tantrums emerge in early childhood and persist into adulthood. They are followed by the appearance of other behavior problems in middle childhood. These include manifestations of cognitive rigidity (e.g., repetitive asking, insistence on sameness, selective attention, difficulty with transitions), and anxiety (e.g., excessive and repetitive actions, emotional lability, and stress sensitivity) [6]. These characteristics define the behavioral phenotype of PWS and are often a focus of treatment from mental health professionals. Behavior management, psychological therapies and psychotropic medications are often prescribed. Symptoms of co-morbid psychiatric conditions (e.g., mood disorder and psychosis) often emerge in adolescence and require medication management [6]. There are some phenotypic differences that correlate with PWS genetic subtypes. Individuals with the deletion subtype are generally more affected with compulsions and self-injury and perform more poorly on cognitive and behavior instruments [7,8]. Those with UPD have higher verbal IQs than those with deletion, but they may be more prone to symptoms of autism spectrum disorder in early childhood and affective psychosis with onset in adolescence and young adult years [9,10]. There is an increased incidence of psychosis among both subtypes [10,11].
The NIH Rare Disease PWS Consortium Registry tracked 355 patients during clinic visits over ten years and recorded age of onset of use of psychotropic medication and pattern of sustained utilization [12]. A total of 265 patients were receiving 483 psychotropic medications. From these data, it appears that multiple medications were used for management of the complex symptoms and co-morbid conditions associated with PWS [13]. Selective serotonin reuptake inhibitors (SSRIs) were used in nearly 50% of the 5–12 year age group and in 70% of the 12–21 year age cohort. Atypical neuroleptics were the second most frequent class of medications used (34%), often in combination with SSRIs [13]. Polypharmacy, the use of more than one medication of a single class or multiple medications from different classes, increases the risk for drug-drug interactions and adverse effects, and this tendency is exacerbated at younger ages. Careful medication selection and informed medication management is required [14].
1.2. Pharmacogenetics of the Cytochrome P450 System
Pharmacogenetics examines the influence of DNA structural variations on genes coding for enzymes responsible for drug metabolism determining efficacy and tolerability. These protein-coding genes are diverse across the genome. The cytochrome P450 enzyme system (CYP) is a heme-based superfamily of proteins responsible for the oxidative phosphorylation of toxins and medications and the synthesis of lipids, steroids (hormones) and some vitamins [14]. This enzyme system is present in most body tissues including the liver and brain [15,16]. It is primarily positioned within the inner mitochondrial membrane or endoplasmic reticulum of the cell. Up to 80% of all drugs are metabolized in the liver by these six different cytochrome P450 enzymes: CYP1A2, CYP2B6, CYPC19, CYP2D6, CYP3A4 and CYP3A5 [17]. CYPD26 by itself may account for the breakdown of up to 25% of all medications [14]. Gene variants have clinically relevant impact on drug metabolism, drug efficacy, side effects, and drug-drug interaction in the clinical setting; they are also associated with susceptibility to cancer and disease [18]. Drugs undergoing metabolism often involve more than one cytochrome enzyme. This is graphically represented for most drugs [https://www.pharmgkb.org/pathways]. In addition, drugs may require first pass metabolism to generate the therapeutic agent for treatment effect. Also, cytochrome P450 genes and their encoded enzymes may be altered by the environment through inhibitors and inducers [14]. The chromosome location for genes encoding cytochrome P450 enzymes that are discussed in this article, their common polymorphisms, and phenotypic activity can be found in Table A1. The Variation (PharmVar) Consortium: Incorporation of the Human Cytochrome P450 (CYP) Allele Nomenclature Database [https://www.pharmgkb.org] is a helpful reference to understand and decode gene notation, population frequency and phenotypic expression.
Compared to all medications, psychotropic drugs are more selectively metabolized by cytochromes 2D6, 2B6, 3A4, 1A2, 2C19 and 2C9 [19]. These genes have many polymorphisms that produce enzymes of variable activity. Some alleles have enhanced activity while others are reduced or inactive. The cytochrome P450 phenotype is defined by the number and combination of alleles inherited from the parents. Gene function is described by four phenotypic categories: ultra-rapid metabolizer (UM); extensive metabolizer (EM); intermediate metabolizer (IM); and poor metabolizer (PM) for each cytochrome P450 gene. The typical or normal rate of activity is the extensive metabolizing phenotype.
Also, gene polymorphisms vary according to race, ethnicity, and geographical ancestry. A comprehensive list of gene polymorphisms and their phenotypic activity are reported at [https://www.pharmgkb.org]. Similar data, together with allelic frequencies and their racial and ethno-geographic frequency, can be found in the review article by Zhou et al. [20].
More than ten years ago at Vanderbilt University, a survey was completed by the parents or caregivers of 86 persons with PWS to ascertain the respondent’s satisfaction with the use of SSRIs and/or atypical neuroleptics to manage behavioral symptoms related to PWS [21]. SSRIs were used in 33%, atypical neuroleptics were used in 11%, and a combination of SSRIs plus atypical antipsychotics was used in 17% of the study population. PWS genetic subtype was not specified. In this pilot study, research-based probes were used to identify phenotypes of CYP2D6 and CYP2C19; allelic frequencies were not reported. Although the EM phenotype predominated for cytochromes 2D6 and 2C19, 37% were IMs of CYP2D6, 2.5% were PMs of CYP2D6, and 3.2% were PMs of CYP2C19. The results of the Vanderbilt study indicated that in more than one-third of persons with PWS, the IM or PM status was noted for CYP2D6. It can be inferred that serum levels of antidepressant or antipsychotic medication metabolized by CYP2D6 would be higher than expected at typical doses, leading to clinical response at lower doses or adverse effects at typical doses. This finding is corroborated by clinical experience with SSRI medications in PWS, as described in the mood and behavioral activation case series reported by Durette et al. [22].
The aim of this clinical report was to examine and summarize pharmacogenetic results from a cohort of patients with PWS referred for evaluation and treatment of psychiatric and behavioral problems. Our approach was to identify differences in CYP genotypes and phenotypes in the referred cohort as a whole, correlate these findings with phenotypic frequencies among the PWS genetic subtypes, and compare these results to data from a normative population. These findings may inform our understanding of why many of our patients have a therapeutic response at low doses of SSRIs and adverse events at typical doses of psychotropic medications as reported by Gourash et al. [23]. This knowledge will be used in our daily clinical practice to guide selection and dose of psychotropic medication, to anticipate potential drug interactions, and to foresee vulnerability to adverse effects while treating the psychiatric and behavioral problems occurring in our patients with this rare genetic syndrome.
2. Methods
Thirty-five patients with PWS (14 DEL; 16 UPD; 5 Unspec-methylation positive, molecular subtype unspecified) were seen for evaluation and treatment at one of three regional centers by the physician authors who have extensive experience using psychotropic medications to manage psychiatric or behavior problems in patients with PWS. The clinical use of pharmacogenetic testing was discussed with and approved by the patient and/or guardian, and testing was ordered to determine cytochrome P450 function to guide the selection of medication and the dosage required for treatment. The authors collected buccal cells in the clinical setting and sent them to one of three CLIA approved and accredited, commercial laboratories: Genesight (GS) in Mason, Ohio (n = 29); Genelex (GL) in Seattle, Washington (n = 5); and Genomind (GM) in King of Prussia, Pennsylvania (n = 1). These laboratories undertake quality control testing required for accreditation to assure and maintain accuracy of laboratory testing results. DNA was extracted at the laboratories and analyzed for polymorphisms of CYP2D6, CYP2B6, CYP2C19, CYP2C9, CYP3A4 and CYP1A2 as well as other genes not included in this case report. Results were received by the authors and protected health information was removed (name, age, gender, race, ethnicity) prior to data pooling and sorting by genetic subtype of PWS. Psychiatric diagnosis, psychotropic medication history, and family psychiatric and treatment history were not available, although all the patients met medical necessity criteria indicating a failure of previous medications, either due to inefficacy or adverse effects, and/or the presence of psychiatric co-morbidities that would require treatment with multiple psychotropic medications.
When comparing the testing results from the three commercial laboratories, there were subtle differences noted. In some cases, different nomenclature was used for the same results, i.e., CYP1A2 *1 and *1A are both names for the wildtype gene, and CYP1A2*1F is the hyper-inducible-163 A/A polymorphism. Also, we found that the interpretation of phenotype from genotype may differ across the commercial laboratories, as each of them uses a combinatorial phenotype that is determined by a proprietary algorithm. For example, GL identifies any carrier of CYP1A2*1F as HI (hyper-inducible), while GS identifies *1F heterozygotes as UM (ultrarapid metabolizers) and GM identifies them as EM (extensive metabolizers). Although we used the phenotypic nomenclature reported by the commercial companies designating metabolizing status as poor, intermediate, extensive and ultrarapid, we also analyzed the frequencies of alleles and diplotypes and compared them to published norms.
For CYP2C19, the pharmacogenomic companies reported CYP2C19 phenotype as ultrarapid, extensive, intermediate, and poor metabolizing. However, the authors were aware that the current phenotypic nomenclature for CYP2C19 has been changed to ultrarapid, rapid, normal, intermediate and poor metabolizing, as described at https://www.pharmgkb.org/page/cyp2c19RefMaterials, to more accurately describe the function of the CYP2C19*17 rapid metabolizing allele. To calculate the phenotypic frequencies for CYP2C19 from the normative data, we combined the frequencies for rapid and normal metabolizing phenotype and used this value for the normative extensive metabolizers, as discussed and implemented in Martis et al. [24]. Pharmacokinetic testing has indicated that there is minimal variance between the normal and rapid metabolizing phenotypes [25].
In this clinical case report, the frequency of phenotypes assigned by the commercial providers for each cytochrome P450 gene is displayed as a histogram for each PWS genetic subtype (DEL, UPD and PWS Unspec) and compared to frequencies found in the normative Caucasian population. Then, for each genetic subtype of PWS, the frequency of phenotypes for each CYP gene was calculated, compared to published normative data, and analyzed for significance using the chi-square test. p values of <0.05 were statistically significant. Next, for each cytochrome P450 polymorphism, the frequency of occurrence of the alleles and diplotypes was calculated for each genetic subtype of PWS and displayed graphically for comparison with normative population data.
For this case series normative data from the Caucasian population was referenced because the NIH PWS Registry found minimal racial/ethnic diversity (93% Caucasian) among 355 enrollees from regional clinics across the USA. The phenotypic frequency in the Caucasian population was obtained from https://www.pharmgkb.org for cytochromes 2D6, 2B6, 2C19, and 2C9, and Zhou et al. was used for 3A4 [20]. The values reported in the literature for the frequencies of CYP phenotypes for each cytochrome show some variability across studies even within designated ethnic categories. Normative data for the phenotypic frequency of 1A2 among Caucasians was not available, although the increased prevalence of the poor metabolizer phenotype among the Asian population is well documented in the literature. We elected to use the results of studies measuring urinary caffeine metabolites to report the normative metabolic phenotype of CYP1A2, a method that has been used for over 20 years and more recently correlated with genotype [26,27,28].
The patients in this clinical case series sought care at one of three specialty programs across the USA. They were evaluated by physician experts with over 80 years of collective clinical experience in treating patients with this rare disorder. These patients received pharmacogenetic testing as part of their medical evaluation because it was deemed as medically necessary. The clinical care described in this article was not part of a research project, so this report did not require ethical review or approval. Prior to the collation and analysis of data, all pertinent private or protected health information about each patient was eliminated or deidentified, except for the genetic subtype of PWS and the pharmacogenetic genotypes and phenotypes. The results of the analysis of this group data will not affect the patient’s clinical care, nor does it have the potential to cause the patient any harm. It is the authors hope that this report will inform, improve, and advance the quality of medical and psychiatric care of patients with PWS.
3. Results
This case series represents the largest number of patients with known genetic subtype of PWS to receive pharmacogenetic testing with analysis of results that are summarized in Table 1. The frequencies of CYP phenotypes in the PWS cohort are itemized for each genetic subtype and compared to a normative (typical) Caucasian population. The list of substrates affected by these CYP phenotypic differences was derived from the frequency of medication use among 265 patients with PWS enrolled in the NIH PWS Registry [12,13]. The raw data obtained from the pharmacogenetic testing of this cohort of referred patients is displayed in the Appendix A. For each genetic subtype of PWS, the testing laboratory, cytochrome P450 genotype, and cytochrome P450 phenotype are specified in a series of tables: Deletion (Table A2), UPD (Table A3) and PWS Unspecified (Table A4). Table A5 shows the proportion of phenotypes for each cytochrome P450 gene according to the PWS genetic subtype.
3.1. Cytochrome P450 Genotypes, Phenotypes and Genetic Subtype of PWS
The cytochrome P450 phenotypes for the combined group of patients with PWS in our referred cohort is shown in Figure 1. The distribution of frequencies of cytochrome P450 phenotypes is displayed as a percentage of the total cohort (n = 35) for ease of comparison with normative data.
The normative data delineating the phenotypic frequencies for CYP1A2 activity (slow, intermediate and rapid) were derived from studies measuring caffeine and its metabolites in urine [27]. Al-Ahmad et al. have described the correlation between metabolic phenotypes and genotypes, e.g., rapid metabolizer phenotype corresponds to *1A/*1A (extensive metabolizer) genotype [28].
Across the combined cohort, the extensive metabolizing status prevailed in all but one cytochrome; for CYP1A2, the ultra-rapid phenotype was more common than the extensive metabolizing. Extensive metabolizing phenotype for the predicted normative data exceeded the PWS cohort for CYP2D6 and CYP2C9 but not for CYP2B6 or CYP2C19. Poor metabolizing status for CYP2D6 and intermediate metabolizing status for CYP2C9 were greater in the PWS cohort than predicted in normative populations. The following series of histograms display the cytochrome P450 enzyme phenotypes for each PWS genetic subtype (Figure 2—DEL, Figure 3—UPD, and Figure 4—PWS Unspecified) and compare these to the predicted normative data for CYP2D6, CYP2B6, CYP2C19 and CYP2C9.
When the data sets for PWS genetic subtypes in our referred population were compared to each other, differences were found in the distribution of phenotypes for all cytochromes. In the following sections, the allelic frequencies of the cytochrome P450 gene polymorphisms are displayed for the PWS genetic subtypes. Also, when possible, the distribution of diplotypes is itemized and compared with normative data. The phenotypic action of the most common alleles for each cytochrome P450 gene is itemized in Table A1 (Appendix A).
3.2. CYP2D6
For CYP2D6 the data from our case report and the Vanderbilt study is compared to the normative American population referenced at [www.pharmgkb.org/page/cyp2b6RefMaterials]. In our combined cohort of referred patients, 48.6% of had the extensive metabolizing phenotype compared to 63.6% among the normative American population; this was not significantly different by chi-square test (p > 0.05). The percentage of CYP2D6 intermediate metabolizers was 34%, and this is similar to the percentage reported in the Vanderbilt survey (37%); both values exceed the 23.6% found in the normative American population, but the chi-square value was not significant (p > 0.05). There were 17.1% poor metabolizers in the current cohort compared to 2.5% in the Vanderbilt survey and 2.2% in the normative American population, and the chi-square value was significant (p < 0.05). When comparing the current data to the Vanderbilt survey, it should be noted that the current cohort was derived from a clinically referred sample, where medical necessity dictated testing. There may have been more treatment failures or adverse events in the current cohort. Data from our referred cohort indicates that over half of the patients with PWS had the intermediate or poor metabolizing phenotype of CYP2D6, which could impact efficacy and tolerability of many of the psychotropic drugs used in treating patients with PWS.
When considering the PWS genetic subtypes, there were fewer extensive metabolizers, more intermediate metabolizers, and more poor metabolizers among those with DEL compared to the normative American population, and all values were statistically significant (p < 0.05). Among those with UPD, the ratio of EM:IM was roughly the same as in the normative American population, although the number of poor metabolizers was actually greater than among DEL, and both PM values were statistically significant (p < 0.05). Figure 5 displays the allelic distribution and frequencies of CYP2D6 polymorphisms among PWS genetic subtypes in our referred cohort.
In our cohort of patients, there is a lesser frequency than predicted for the most common CYP2D6 alleles *1 and *2 (both convey normal activity) and an increased frequency of alleles *4 (inactive) and *41 (reduced function). Further, the distribution of alleles includes others with a lower frequency of occurrence that are inactive or have reduced function. Subtle differences were noted in the number of alleles between the PWS genetic subtypes, but the DEL group had a higher frequency of *4 alleles, and the UPD group had a greater number of *41 alleles. See Figure 6 for CYP2D6 diplotypes.
Among the total cohort of PWS, the wild type diplotype *1*1 occurs at a reduced rate, roughly two-thirds of the American normative population, but nearly equal to the frequency of the *1*4 diplotype, which codes for decreased activity, and exceeds the normative frequency by more than one-third. The most frequent CYP2D6 diplotypes among the DEL subtype are *1*4 and *1*9, both of which have decreased activity predicting intermediate metabolizer phenotype. The frequency of *1*1, which is the extensive metabolizing wild type, is equal in frequency to *1*9, which has decreased activity. These frequencies explain the predominance of intermediate metabolizer status among DEL. Among the UPD subtype, the highest frequencies are *1*2A (extensive metabolizing) and *1*4 (intermediate metabolizing), and the next most frequent are *1*41 (extensive metabolizing) and *4*41 (poor metabolizing), explaining the 9EM:4IM:3PM ratio of metabolizer phenotypes.
Because CYP2D6 metabolizes many antidepressants and antipsychotics often prescribed in PWS, it is not a surprise that our cohort of patients referred for treatment has alleles with reduced function contributing to decreased efficacy or adverse effects; conversely, they may respond to a lower dose [29,30]. Medications (in alphabetical order) most commonly used in PWS that are substrates of CYP2D6 include amphetamines, aripiprazole, bupropion, citalopram, clonidine, diphenhydramine, escitalopram, fluoxetine, fluvoxamine, haloperidol, olanzapine, quetiapine, risperidone, sertraline, trazadone and ziprasidone [31]. The CYP2D6 enzyme activity in our cohort showed a greater number of patients than predicted with poor metabolizer status among both PWS genetic subtypes, which could impact approximately 25% of all medications and 60–70% of behavioral/psychiatric prescribed drugs as discussed by Butler [18].
3.3. CYP2B6
Among our total PWS cohort, 53% had the extensive metabolizer phenotype compared to 43% in a normative European population [www.pharmgkb.org/page/cyp2b6RefMaterials]. The chi-square test was not significant (p > 0.05). Intermediate metabolizers were found in 47% of the patients with PWS compared with 39% in the European population, but again, the chi-square value was not significant (p > 0.05). There were no poor metabolizers among our patients with PWS. The intermediate metabolizing phenotype predominated among the UPD cohort, whereas the intermediate and extensive metabolizing phenotypes were equal among DEL. These results were not statistically significant by chi-square test (p > 0.05). The distribution of alleles for CYP2B6 is shown in Figure 7.
There were 4 alleles identified in our cohort with a greater expression of the wild type allele (CYP2B6*1) that confers normal activity in the combined cohort (69%) compared to the normative European population (47%). In the combined cohort the expression of *6, an allele associated with decreased function, is nearly equal to the normative population. Comparing DEL and UPD genetic subtypes, there is a greater expression of *6 allele among the UPD cohort.
Differences in the distribution of diplotypes are noted among the PWS genetic subtypes. Among both DEL and UPD, the frequency of the CYP2B6*1*6 diplotype, which is associated with the intermediate phenotype, is expressed more frequently than the *1*1 diplotype, which has the extensive metabolizer phenotype. In the UPD cohort, *1*6 is expressed almost twice as frequently as *1*1, and this explains the greater number of intermediate metabolizers among the UPD group. The diplotype CYP2B6*1*5, which is associated with normal function, occurs more frequently among DEL than the normative population, but is not found at all among UPD.
Among the total cohort of PWS, both the *1*1 and *1*6 diplotypes are expressed at nearly twice the frequency of the European normative population. This explains the distribution of phenotypic activity among our cohort where the extensive and intermediate metabolizer status are nearly equal. These results would suggest caution when prescribing medications metabolized by CYP2B6, such as bupropion and sertraline among the UPD group.
3.4. CYP2C19
Eighty-six percent of the PWS combined cohort had the extensive metabolizer phenotype of CYP2C19, and this is higher than expected from American normative data (76.4%) but not significantly different by chi-square test (p > 0.05). Less than 10% in the current study were intermediate metabolizers compared to 21.4% for the American population, which again was not significant by chi-square (p > 0.05). Ultra-rapid metabolizers were seen in 5.7% of the patients with PWS compared to 0.7% for the American data, which was not significant by chi-square (p > 0.05). There were no poor metabolizers among the current cohort compared to 3.2% in the Vanderbilt survey. For CYPC19, there were PWS subtype group differences. Among those with deletion, there were more ultra-rapid metabolizers compared to the normative population (p < 0.05). The UPD cohort displayed 100% extensive metabolizer phenotype and no intermediate metabolizers, and both results were statistically significant (p < 0.05) compared to the normative population of 76.4% and 21.4%, respectively. The frequency of alleles and diplotypes are compared with normative American data in Figure 8.
Among our cohort of referred patients with PWS, there is a lesser frequency of the most common CYP2C19 alleles *1 and *2 that have normal function and an increased frequency of *17 allele, which is associated with increased function; there was one person in the PWS Unspecified diagnostic group with *8 allele (inactive). Differences were noted between the PWS genetic subtypes with the *17 allele being highest among DEL. The allelic frequency of *17 in our combined cohort was more than twice that predicted in the American population (8.6%), and three times higher among the DEL. This is the reason for the increased frequency of ultra-rapid metabolizers among DEL and among the total cohort as well.
The frequency of the normal functioning diplotype *1*1 is less than predicted by normative data except among the UPD group. Among the UPD cohort, only extensive metabolizers were found. The number with increased function *1*17 is more frequent due to the presence of *17 allele, especially among DEL. Also, the ultra-rapid functioning diplotype *17*17 is noted among DEL but not UPD. Overall, the extensive metabolizer phenotype of CYP2C19 prevails in our PWS cohort largely due to the increased expression of the *1 and *17 alleles.
Many psychotropic medications used in patients with PWS are substrates for CYP2C19, including (but not limited to) citalopram, clomipramine, doxepin, escitalopram, fluoxetine, imipramine, and sertraline [32]. It is inferred that these medications would have been well tolerated by most of our referred cohort. However, gonadal steroids (estradiol and testosterone), which are replaced commonly in PWS due to delayed or absent puberty, are substrates for CYP2C19, indicating a potential for drug interactions. This is discussed more fully in Section 4.1 and Section 4.2.
3.5. CYP2C9
The prevailing phenotype of CYP2C9 in our cohort was extensive metabolizing (61.8%), which was decreased in comparison to American normative data (83.2%) and was significantly different by chi-square test, p < 0.05. The intermediate metabolizing phenotype was found in 35% of the combined cohort, and this is twice the number predicted in the American population (16.4%) and significantly different also by chi-square test (p < 0.05). For CYP2C9 the DEL cohort was 76.9% extensive metabolizer and 23.1% for intermediate, remarkably similar to American normative data at 83.2% and 16.4%, respectively. But for the UPD cohort, extensive metabolizer was 56.3%, intermediate 37.5%, and poor 6.3%; all of these values were statistically significant by chi-square (p < 0.05).
Among our cohort of patients with PWS, the distribution of alleles in Figure 9 shows a predominance of CYP2C9*1, which confers normal activity at a lesser frequency than in the normative American population. There is a greater occurrence of the CYP2C9*2 allele (16.2% of ALL PWS) compared to the frequency in the American population (3.3%) with greatest prevalence among UPD (18.8%). Alleles *2 and *3 have decreased activity, and there were no CYP2C9*3 alleles among DEL. Among the UPD cohort, the frequency of CYP2C9*1 allele was less than DEL, and the CYP2C9*2 allele frequency was nearly twice that found among DEL.
The CYP2C9*1*1 diplotype that confers the normal phenotype has the greatest frequency across all PWS genetic subtypes, although it occurs more frequently among DEL than UPD. All other pairs are intermediate metabolizing, and there is a greater proportion of these pairs in UPD compared to DEL. This likely contributes to the greater number of intermediate metabolizers among UPD, twice as many as DEL. More than half of patients with UPD in this cohort may have required dosage adjustment for drugs such as amitriptyline, fluoxetine, and sertraline especially when used with concurrent oral contraceptives, methylphenidate, modafinil and omeprazole [31].
3.6. CYP3A4
In the admixed American population, 97.3% had the extensive metabolizing phenotype of CYP3A4 [20]. Across all PWS genetic subtypes the predominate phenotype of CYP3A4 was extensive metabolizing; there were only 5 patients who had intermediate metabolizing status. CYP3A4 has over 30 polymorphisms, most of which occur at low allelic frequencies. The distribution of alleles and diplotypes is found in Figure 10.
There were only three alleles present in the analysis of CYP3A4: *1A, *22, and *1B across PWS genetic subtypes in our cohort of referred patients. The wildtype gene, CYP3A4*1A, was the major allele expressed with a frequency of 92.6% that compared favorably with Caucasian population norms of 92.1% [33]. CYP3A4*22 is a reduced function allele. The allelic frequency for CYP3A4*22 in this study was 5.9%, and this was consistent with population norms of 5–7% [35]. CYP3A4*1B was expressed only among the PWS UPD subtype at a frequency of 3.3%, which is less than allele frequencies of 7.9% reported among Caucasians [33]. This value was not statistically significant by chi-square, p > 0.05. CYP3A4*1B has been associated with patients who have cancer, and recently Swiechowski et al. found that compared to controls, patients suffering from recurrent major depressive disorder were more likely to have the heterozygous (AG) CYP3A4*1B genotype [36]. In this study comparing 102 patients with 90 controls, the G allele frequency was higher among patients than controls, and those with the homogeneous (GG) genotype, although fewer in number, reported an earlier age of onset of depression. Even though CYP3A4 is involved in the metabolism of many antidepressant medications, such as tricyclics, SSRIs, SNRIs, and mirtazapine, the phenotypic results did not support any differences in metabolism [36].
The distribution of diplotypes finds that the frequency of the wildtype alleles *1A*1A predominates across all PWS genetic subtypes at nearly typical frequency. The occurrence of the reduced function *1*22 diplotype is greater than predicted among all PWS genetic subtypes [34].
3.7. CYP1A2
For CYP1A2, the frequency of ultra-rapid metabolizing phenotype exceeded the extensive metabolizing among the total cohort, and this most likely reflects the influence of the 2:1 frequency of ultra-rapid metabolizers to extensive metabolizers among patients with UPD. Further there is a low frequency of intermediate metabolizers, and there are no poor metabolizers. Compared to other cytochromes, CYP1A2 appears to be unique in the number of polymorphisms, the variety of inherited allelic combinations, and the capacity for induction from medicinal, dietary, gender, hormonal, and lifestyle factors [14].
Across all PWS subtypes, there was an increase in the number of ultrarapid metabolizers, especially among UPD (p < 0.05), and there were no poor metabolizers across the total PWS cohort compared to the normative population (p < 0.05). A significant difference was noted by chi-square test (p < 0.05) when comparing the frequency of CYP1A2 IM metabolic phenotype to the normative population (54%) in our patients with DEL (0%) and UPD (6.25%) [27].
There were six alleles present in the analysis of CYP1A2 in our cohort (*1A, *1B, *1C, *1D, *1E, and *1F) and their frequencies among each genetic subtype are presented in Figure 11 and compared to predicted values among Caucasians [37,38].
CYP1A2*1A and *1B have normal function and CYP1A2*1C has decreased function. Both CYP1A2*1D and CYP1A2*1F are inducible. For example, both CYP1A2 *1A and *1F express the extensive metabolizer phenotype in the absence of an inducer, but in the presence of tobacco smoke, insulin, modafinil, nafcillin, omeprazole or cruciferous vegetables, the hyper-inducible (HI) phenotype is expressed by *1F [38]. Among the pharmacogenetic results from the Genelex company, the *1F haplotype was identified as having the HI phenotype. To achieve consistency across the data set for CYP1A2 in Table 1, we combined the results of HI into EM.
The gene for CYP1A2 is located on chromosome 15 (15q24.1) outside the PWS critical region. Therefore, it is possible that in the PWS patient with maternal uniparental disomy 15 (UPD), the CYP1A2 alleles may be identical. The clinical relevance of this depends upon the phenotypic activity determined by the CYP1A2 alleles seen in the mother. Almost all people have 2 copies of the CYP1A2 gene [38]. Among our cohort of patients with UPD, there were 3 duplications of alleles, whereas cohorts with DEL and PWS Unspecified had only one. Also, there were complex genotypes of CYP1A2, some with as many as 4 alleles. Clinical implications for treatment with these CYP1A2 findings in PWS are discussed in Section 4.1 and Section 4.2.
4. Discussion
The use of pharmacogenomic testing is now commonly obtained, particularly after an untoward result of a medication treatment trial. It provides guidance for medication selection and dosing. There are several studies suggesting that genotypically informed medication selection for treatment of depression increases response and remission rates, decreases adverse effects, and guides the use of adjunctive medications in challenging cases [39,40].
Knowledge of the pharmacogenetic phenotype can inform time parameters of treatment response. Ultra-rapid metabolizers are more likely to have a quicker, positive clinical response to antidepressant medication with several studies suggesting an increased risk of rehospitalization and emergency room visits [41]. In a study performed at the Institute of Living, those patients with the CYP2D6 ultra-rapid metabolizing genotype were more likely to be discharged early and to have at least one readmission within the month after discharge [42]. On the other hand, those patients with intermediate or slow metabolizing genotypes were likely to take a longer time to achieve a clinical treatment response, a longer duration of hospital stay, and a decreased likelihood of readmission in the 30 days after discharge [42]. In another study, intermediate metabolizers were more likely to have adverse effects to medications metabolized by CYP2D6 than extensive metabolizers at comparable doses [43].
Knowledge of the pharmacogenetic phenotype can inform dose response parameters [44]. For example, if a person has the poor metabolizing phenotype of CYP2D6, which metabolizes paroxetine exclusively, a typical starting dose of medication may produce a therapeutic effect, and according to Kirchheiner et al. the dose required for remission may be 65% of the standard dose [30]. If the person is an intermediate metabolizer, they may tolerate low to moderate doses of medication, but as the dose is titrated, their metabolic capacity may be exceeded resulting in side effects [45]. On the other hand, if a person is an ultra-rapid metabolizer, the half-life of the medication will be reduced and the patient may experience unpleasant withdrawal symptoms during the day [29], such as headache and gastrointestinal upset without fever, mood instability, and what is been described as the perception of a lightning bolts radiating down the arms. With this ultra-rapid genotype, these symptoms can be addressed by increasing the schedule of administration of medication across the day. As a result, and following the standard of care, the patient may require a higher total daily dose, e.g., for paroxetine, 135% of the standard dose, than typically recommended by the FDA for that medication and condition as discussed in Kirchheiner et al. [30].
Knowledge of the pharmacogenetic phenotypes can inform other parameters of medication response, including potential adverse effects. If a medication requires conversion to an active metabolite by the cytochrome enzyme system, the phenotype will determine the rate of activation. The best example of this is codeine, which requires CYP2D6 for conversion to the active metabolite morphine. Poor metabolizers of CYP2D6 are at risk for a poor analgesic response, as ultra-rapid metabolizers are at risk for toxicity [29]. Another example is risperidone, one of the most widely prescribed second generation antipsychotic medications among children and adolescents, including with PWS, with known efficacy and adverse effects, especially among those with autism spectrum disorder (ASD) and intellectual and developmental disabilities (IDD) [46]. Risperidone is primarily metabolized by CYP2D6 into another active metabolite, paliperidone, which is linearly related to serum prolactin level. Individuals with the ultra-rapid metabolizing phenotype of CYP2D6 may be more likely to display hyperprolactinemia [47]. In a study of 257 children and adolescents who received risperidone, 76 experienced a variety of adverse effects, and these were more commonly seen in the poor or intermediate metabolizing phenotype of CYP2D6 [48].
Knowledge of potential drug interactions can inform treatment and minimize adverse effects. Drug interactions can occur when using prescribed medications as well as over the counter (OTC) agents and nutraceuticals (herbs, supplements, or vitamins) [14].
4.1. Drug–Drug Interactions
Drug-drug interactions are a common cause of adverse events or failed treatment efficacy. Checking for potential drug interactions can be accomplished by examining the pharmacokinetic pathways for medication metabolism at https://www.pharmgkb.org/pathways and https://drug-interactions.medicine.iu.edu [32]. The Flockhart table is updated frequently and delineates the pharmacogenetic action of specific drugs that use the cytochrome P450 enzyme system. Drugs are delineated as substrates, inhibitors, or inducers for one or more cytochrome P450 enzymes. Medications administered concurrently that are substrates for the same cytochrome enzyme overwhelm the metabolic capacity of the cytochrome and interfere with efficacy or may result in toxicity [45]. If two drugs have the same affinity for an enzyme, they become competitive inhibitors in a dose related way [49]. If one of the drugs is a substrate for another cytochrome P450 enzyme with less affinity, metabolism may be shunted toward those less preferred pathways. An inducer is a drug or agent (e.g., cigarette smoking) that causes an increase in the production of the cytochrome enzyme by action at the promoter site on the gene and usually takes 1–2 weeks to occur [39,40]. An inhibitor is a drug or agent (e.g., cruciferous vegetables) that binds to the cytochrome enzyme and blocks its use; its effect is immediate and persistent for as long as the inhibitor remains in the system. Some medications act as both substrate and inhibitor (e.g., fluoxetine with CYP2D6), or substrate and inducer (e.g., carbamazepine with CYP3A4) [35].
Phenoconversion describes the process whereby a given CYP phenotype, inferred from genotype, is functionally converted to a higher or lower metabolic status on the basis of drug-drug interactions or the influence of non-genetic factors [50]. The resulting phenotype carries the designation of phenocopy because it imitates a different metabolic status. Phenoconversion is more likely to occur in drug-drug interactions when a substrate has a high affinity for a single CYP. For example, clozapine and olanzapine are selectively metabolized by CYP1A2; in the presence of cigarette smoking, phenoconversion to a higher metabolizing phenotype results in the reduction of serum drug levels by 20% [43]. In the case of aripiprazole, concurrent administration of a CYP2D6 inhibitor, such as bupropion or sertraline, can convert the 2D6 extensive metabolizing status downward, driving serum drug levels upward by 20–50%, precipitating adverse effects [51]. In another example, ratios of metabolic phenotypes that were inferred from genotype across cytochromes 2C9, 2C19, 2D6 and 3A4 were converted to a lower status in the presence of inflammation, as reviewed by Klomp et al. [50].
Table 2 itemizes psychotropic and other prescribed medications, OTC agents, nutraceuticals, and dietary/lifestyle factors that are substrates, inhibitors, or inducers of cytochrome P450 enzymes. Most of the items in this table were generated from clinical experience with persons who have PWS and by referencing Flockhart [32]. This table does not constitute a complete list of medications, nor is it a compendium of recommended therapeutic agents for patients with PWS.
Drug–drug interactions may occur with prescribed medications as well as OTC agents and nutraceuticals. OTC antihistamines, such as loratadine (substrate of CYP3A4), may interfere with metabolism of some antidepressants and antibiotics [52]. Chlorpheniramine and dextromethorphan, OTC agents used for common cold symptoms, are substrates for CYP2D6 [32]. Competition at the site of the enzyme receptor can displace risperidone and potentiate its efficacy producing increased sedation or extrapyramidal side effects such as akathisia. The anti-acid medication ranitidine, which was recently removed from the market, is a substrate for CYP2D6 and competes with several atypical antipsychotic medications, including risperidone [18]. Toxic side effects have been reported, including incapacitating sedation that was misdiagnosed as dementia in a patient with PWS. Similarly, omeprazole is a substrate for CYP2C19 and CYP2C9 and inhibits metabolism of many antidepressants contributing to the potential toxicity. Acetaminophen, a substrate for CYP1A2, and ibuprofen, a substrate for CYP2C9, may affect efficacy of psychotropic medications administered concurrently [38]. Melatonin, a substrate of CYP1A2, is commonly used for treatment of insomnia, and treatment with fluvoxamine results in increased serum melatonin [53].
Herbal or plant-based agents may also interact with psychotropic medications. Ginkgo biloba, used to treat memory and dyskinesias, has been associated with increased risk of hemorrhage in combination with SSRIs or SNRIs [54]. Ginseng, used to increase energy, stamina and well-being, has been associated with serotonin syndrome in combination with SSRIs or SNRIs, and ventricular arrythmias in combination with haloperidol [54]. Goldenseal, used for upper respiratory and gastrointestinal tract symptoms, is a potent inhibitor of CYP3A and CYP2D6 [53]. Milk thistle, used for diabetes or liver disease, has resulted in pancreatitis when co-administered with haloperidol or risperidone, and hepatotoxicity may occur with aripiprazole [54]. St. John’s Wort, a natural occurring antidepressant, interferes with the efficacy of many pharmaceuticals due to induction of cytochromes 1A2, 2C9 and 3A4 [55]. Both cannabis and cannabidiol are substrates for CYP2C19 and CYP3A4 and inhibitors of CYP2D6 and CYPC9; also, cannabis is a substrate for CYP2C9, and cannabidiol is an inhibitor of CYP2C19 [56]. Their concurrent use with fluoxetine, a substrate for CYP2D6, CYP2C9, and CYP2C19 and an inhibitor of CYP2C9 and CYP2C19, can be expected to produce a complex drug interaction that may result in mood activation. Because each of these agents has a relatively long half-life, the onset of any clinically relevant interaction may be delayed, and once it is identified, the discontinuation of one of the medications will take a while to clear the body, so unpleasant symptoms may linger. The persistence of mood and behavioral symptoms may lead to the diagnosis of a co-morbid psychiatric condition or may require management with yet another psychotropic medication.
Another common drug-drug interaction is seen with SSRIs (citalopram, escitalopram and sometimes sertraline) and estradiol and/or progesterone (taken for hormone therapy, oral contraception, or occurring naturally with monthly menstrual cycles), which are substrates for CYP2C19. A woman receiving sertraline for depression who experiences premenstrual dysphoria may benefit from a transient dose increase during the luteal phase as progesterone levels increase [57,58]. Because puberty is often delayed or absent in PWS, hormone therapy is usually initiated during the age of typical adolescence. If SSRIs had been started previously, the addition of estradiol or a combination pill may compete with CYP2C19 metabolism altering serum levels and potentially precipitating mood and behavioral difficulties requiring a dose adjustment. Hormone therapy is likely to continue into adulthood in PWS to address lifelong osteopenia and osteoporosis.
The frequency of polypharmacy use identified in the NIH PWS Registry suggests that drug interactions, like the ones identified above, may be commonplace. Knowledge of pharmacogenetics may inform dose parameters as well as potential drug interactions. For patients with PWS, like others with intellectual and developmental disabilities, it is always advisable to try one medication at a time, or to add another medication after a new behavioral baseline has been established. This applies for hormone therapies and use of nutraceuticals as well.
4.2. Specific Relevance of Cytochrome P450 Enzyme System to PWS
There are factors that can change the phenotypic expression of the cytochrome P450 genes that are particularly relevant to PWS. CYP3A4,5 is sexually dimorphic due to gonadal steroid effects on gene activity. There is evidence to suggest that growth hormone is a modulator of this gender specificity of function. Sinues and colleagues [59] examined CYP3A enzyme activity in 35 unrelated growth hormone deficient children (ages 2.9–13.1 years) both at baseline and after growth hormone replacement. At baseline, the level of activity of CYP3A was elevated compared to controls in a non-sex dependent manner. Then, after growth hormone replacement for 30 days, CYP3A activity was reduced to normal range in males but was unchanged in females. This typical sexual dimorphic level of activity for the CYP3A enzyme impacts serum levels of testosterone, gonadal steroids, as well as psychotropic drugs. This has special relevance for individuals with PWS who are likely to be growth hormone deficient and require growth hormone replacement as well as gonadal steroid therapy [1].
The activity of CYP1A2 has gender specificity also, with lower metabolic function in women than in men. [60,61]. The primary function of CYP1A2 is to purge potential environmental toxins, e.g., heterocyclic amines, and polycyclic hydrocarbons, that may act as carcinogens from the body. Pharmacogenetic studies have explored polymorphisms of these alleles as predisposing factors to the incidence of cancers of the urinary tract, colon, and rectum [37]. CYP1A2 is inhibited by estradiol, oral contraceptives, some antibiotics (ciprofloxacin and levofloxacin), and other drugs (fluvoxamine, celecoxib, and amiodarone). Polymorphisms of CYP1A2 have been explored as predisposing factors for psychiatric illness [62]. Yenilmez et al. found that the allelic frequency of CYP1A2*1F among those with bipolar disorder was 25.5% and among those with schizophrenia, it was 69.4%, which is twice that predicted for normative population frequencies [62]. Because many psychotropic medications used for treating these conditions are also substrates for CYP1A2 (e.g., olanzapine, clozapine, haloperidol, fluvoxamine, duloxetine, tricyclic antidepressants, and clomipramine) [18], their metabolism is subject to the effects of induction, which reduces serum concentration and can interfere with treatment efficacy. Because of hypogonadism and delayed puberty in PWS, females are often prescribed estradiol or estrogen/progesterone combinations found in oral contraceptives, both of which inhibit gene expression and activity of CYP1A2. This inhibition may result in increased drug levels of substrates of CYP1A2 with potential toxicity [32,63].
CYP1A2 is highly inducible by dietary factors and lifestyle considerations [18]. Caffeine is a well-recognized substrate of CYP1A2, and it is used to determine enzyme activity [37]. The CYP1A2*1D and CYP1A2*1F alleles predict the ultra-rapid metabolizing phenotype when the enzymes are induced. Common inducers are tobacco and cannabis smoke, chargrilled meats, cruciferous vegetables (e.g., brussels sprout, broccoli, cauliflower, cabbage, radish, rocket, watercress, and wasabi), insulin, modafinil, nafcillin, omeprazole, and carbamazepine [14,38]. Xie at al. reviewed the magnitude of induction effects, with 3.5-fold increase in smokers vs. non-smokers and 25% increase with broccoli containing diets [63]. The inducibility of CYP1A2 by diet is particularly important to persons with PWS, many of whom follow the Red Yellow Green Diet that contains a high percentage of cruciferous vegetables that are low in calories and high in fiber [64]. Further, many persons with PWS are prescribed modafinil (CYP1A2 inhibitor) for excessive daytime sleepiness. Cigarette smoking and overuse of caffeinated beverages (CYP1A2 inducers) are not uncommon among adults with PWS. Severe skin picking may result in cellulitis that requires systemic antibiotic treatment (CYP1A2 inhibitors) [65]. Gastroesophageal reflux is common in PWS and often treated with proton pump inhibitors such as omeprazole (CYP1A2 inducer) [65,66].
4.3. Non-Genetic Factors Affecting Drug Pharmacokinetics in PWS
There are non-genetic pharmacokinetic factors determining dose response efficacy, and these factors are especially important in the pediatric population. In this report, the cytochrome system pertains to those enzymes in the liver responsible for metabolizing medications. Overall, the metabolic activity of the liver is increased during the developmental years before puberty such that metabolic clearance in children is twice as fast as in adolescence [29]. Therapeutic doses of medications metabolized by CYP1A2, CYP2C9 and CYP3A4 are much higher in children than compared to adults, and the bioavailability of medications that require first pass metabolism for activation is decreased [56]. Increased metabolism suggests a shorter half-life and indicates that the medication should be dosed more frequently through the day. These developmental changes are not seen with CYP2D6 or CYP2C19 [61].
Diseases or nutritional status that affect the liver function will impair metabolism at any age, such as steatohepatitis. In a study by Li et al., CYP2C19 activity was decreased by as much as 80% in adolescents with steatohepatitis (not steatosis) while activity of CYP1A2, CYP2C9 and CYP3A4 was unaltered, unlike the pattern of change in adults resulting in decreased activity of CYP2C19, CYP1A2 and CYP3A4, and increased activity of CYP2C9 [67].
There are other factors affecting drug metabolism such as body weight or body surface area that determine the daily dosage of medication before puberty. As reviewed by Li et al., obesity is associated with an increase in CYP1A2, CYP2C9, CYP2C19, and CYP2D6 activity in children, and a decrease in CYP1A2 and CYP3A4 activity in adults [67]. Body composition is also important, because lipophilic drugs accumulate in adipose tissue and prolong rate of excretion. The mode of delivery of medication (intramuscular, subcutaneous, or intravenous injection; intranasal, sublingual, or rectal administration; or topical application) can affect speed of onset of action, effectiveness on the target symptom, and extent of systemic side effects. Lifestyle factors can also affect cytochrome function, e.g., consumption of chargrilled foods and cruciferous vegetables induces CYP1A2 and CYP3A4 activity; grapefruit juice inhibits CYP3A4; caffeinated beverages inhibit CYP1A2, and smoking cigarettes or cannabis will induce CYP1A2 [60]. Stomach acidity can also affect absorption, and urine pH can affect excretion, especially with amphetamines [68].
Finally, compliance is the most important variable in determining medication efficacy. In general, the more frequently a medication must be administered, the higher the likelihood that doses will be missed. In the experience of PWS experts who administer care to patients with PWS, it is always preferable to start a regular acting medication first to ascertain dose and side effects before converting to a sustained release form (J.F, personal communication; [69]).
4.4. Limitations and Future Directions
The data presented in this clinical case series were obtained from approved commercial laboratories. Not all commercial pharmacogenomic platforms test the same cytochrome P450 genes and polymorphisms. Companies select the polymorphisms to test based on the patient population, the frequency of polymorphisms, and the clinical relevance for a particular field of medicine, e.g., oncology or psychiatry. Also, the clinical interpretation of testing results may differ. The genetic data obtained may be the same, e.g., both the number and type of alleles, but the designated phenotype is determined by combining results from several genes. The algorithm for this combinatorial approach is proprietary [19].
The idea for reporting this clinical case series was conceptualized after the initial collection of data within our respective clinical practices. The patients who received pharmacogenomic testing met medical necessity criteria set forth by the companies based on insurance coverage, including failure of at least one psychiatric medication. Therefore, our cohort represented a biased population of PWS patients who might be expected to have some alterations of cytochrome P450 function. In this case report, there is no PWS control group for comparison. In addition, the information available for each patient came from three different clinical centers, was subject to record availability (age, gender, ethnicity), and was deidentified prior to data collation and analysis. A history of psychotropic medication use as well as family history of psychiatric conditions and medication response was not discoverable. Also, non-genetic, environmental factors that may influence how these patients metabolize psychotropic medications were not known. Our data was limited to the genetic polymorphisms of selected cytochrome P450 enzymes, our interpretation of these findings, and the PWS genetic subtype of the patient. We have provided a descriptive analysis of these results.
This data set was limited in number and did not reach sufficient power for PWS-subtype analysis in all cases. In particular, the numbers were too small to undertake analysis of gene allele frequencies via Hardy-Weinberg equilibrium studies. None the less, we found statistically significant differences in the metabolizing status of cytochromes 2D6, 2C9, 2C19, 3A4 and 1A2. The results from our cohort may not reflect the general PWS patient population, as the patients referred to our specialty centers are often those who have complicated health/behavior/psychiatric concerns and have failed previous medication management. Further, our PWS cohort consisted of more patients with UPD (46%), which is higher than predicted based on studies showing that 35% of PWS individuals have this genetic subtype [1,2,3]. This may reflect that those with UPD may have more behavioral/psychiatric problems and are more likely to seek mental health intervention. In the future, a larger study of at least 100 subjects from the general population of persons with PWS would provide sufficient statistical power to correlate pharmacogenetic data with factors of age, gender, ethnicity, and family psychiatric history. Also, a larger study would offer greater insight and guidance into medication management in this rare population often presenting with significant psychiatric needs. Until a larger study is performed, the authors realize that the interpretation and generalization of these results is limited to our cohort of referred patients.
Except for CYP1A2, which is located on Ch 15, we cannot explain the frequency of phenotypic differences in the cytochrome P450 system noted in this case series. UPD status due to maternal isodisomy 15 may have altered CYP1A2 allele frequencies in our referred cohort [2]. The most common etiology for CYP polymorphism is familial, which includes variations associated with ethnicity. The ethnicity of our 35 patients was deidentified, and we did not have access to familial psychiatric history. However, the NIH PWS Registry enrolled patients from PWS clinics at 4 sites in the USA, and the racial/ethnic diversity of these 355 patients was minimal with 93% Caucasian. As such, our phenotypic comparisons and statistical correlations were all based on normative Caucasian populations. It is highly unlikely that our pharmacogenetic findings in this PWS cohort were due to ethnic variability.
Finally, one of the criteria for obtaining pharmacogenetic testing includes failure to respond to current treatment with psychotropic medications. The PWS phenotype is a complicated mix of anxiety, stress sensitivity, mood lability, and disruptive behavior. Clinicians assume that these symptoms will respond to psychotropic medication. But it is possible that these symptoms require environmental, psychological, or behavioral management strategies for optimal response. Therefore, treatment failure with psychotropic medications alone cannot necessarily be attributed to polymorphisms of the cytochrome enzyme system.
5. Conclusions
This clinical case series of patients with known PWS subtype is the largest reported cohort to receive commercial pharmacogenomic testing. At first glance, the results from our combined cohort suggest that our patients with PWS had a wide array of alleles that were distributed in a manner consistent with natural variation. However, these patients were referred for evaluation and treatment, and not surprisingly, group differences were noted when comparing the results of PWS testing with normative population data. For example, 48.6% of the combined PWS cohort (n = 35) were extensive metabolizers of CYP2D6 compared to the typical population frequency of 63.6% among Americans. There were 36% intermediate metabolizers and 17.1% poor metabolizers compared to 23.6% and 2.2% respectively among Americans, which means that more than half of our PWS cohort had reduced function of CYP2D6. Further, there were differences between the molecular classes of PWS with the deletion subtype having fewer extensive metabolizers and more intermediate metabolizers; both DEL and UPD subtypes had more poor metabolizers compared to the normative American population. These results suggest that increased vigilance is required by clinicians for dose selection and speed of titration of psychotropic medications used in patients with PWS that are substrates for CYP2D6, such as aripiprazole, bupropion, chlorpromazine, citalopram, clonidine, doxepin, escitalopram, fluoxetine, haloperidol, imipramine, mirtazapine, olanzapine, quetiapine, risperidone, sertraline, trazadone, venlafaxine and ziprasidone. The clinical mantra of start low and go slow is recognized and indicated in the treatment of clinical symptoms among patients with PWS as well as other intellectual and developmental disabilities [45]. The increased frequency of ultra-rapid metabolizing status of CYP1A2 in PWS, especially among those with UPD, suggests that careful selection of drug dose is indicated for agents that are substrates, such as acetaminophen, benztropine, chlorpromazine, clomipramine, clozapine, duloxetine, estradiol, fluvoxamine, haloperidol, imipramine, mirtazapine, and olanzapine [32]. Further, growth hormone status and gender may play a role in pharmacokinetic parameters of patients receiving medications that are substrates for CYP3A4 primarily, such as aripiprazole, guanfacine, mirtazapine, modafinil, quetiapine, trazadone, and ziprasidone [32,69]. Also, both estradiol and testosterone are substrates for CYP3A4. Increased expression of the CYP3A4*1B allele was seen only among UPD in this case series. This allele has been associated with risk of prostate cancer, leukemia, early puberty and major depressive illness [64]. Gender may play a role in the dose selection for persons receiving bupropion or sertraline, as the number of CYP2B6 extensive metabolizers may be higher in PWS as a whole and among females with deletion subtype specifically [70]. A larger cohort of non-referred persons with PWS is needed to substantiate these preliminary pharmacogenetic findings and their applications to the clinical setting where patients with this rare disorder are treated.
In conclusion, like any medical test, pharmacogenetics testing requires that medical necessity criteria be met. These criteria include symptom severity, co-morbid illnesses, and number of failed medication treatment responses. These factors increase the likelihood that more than one medication will be used, as is often seen in PWS. Polypharmacy increases the potential for drug interactions. The pharmacogenetic data from this case series of referred PWS patients underscores the benefit of knowing the phenotypic function of the cytochrome P450 enzyme system to inform selection of psychotropic medication and dose to improve the care and treatment of those with this rare genetic condition. Although our cohort was too small to discover a unique pharmacogenetic phenotype among our patients, we did identify some statistically significant differences in phenotypic function compared to population norms and also between PWS genetic subtypes. Going forward, a larger cohort of non-referred persons with PWS must be studied to confirm these observations for widespread clinical application. Nonetheless, this data supports the use of pharmacogenetics testing in Prader-Willi syndrome.
Acknowledgments
We thank Grace Graham for her expert preparation of this manuscript. We also acknowledge Linda Gourash, who is a Pittsburgh Partnership collaborator, Waheeda Hossain for statistical analysis, the patients with PWS and their families along with the Prader-Willi Syndrome Association (USA) for their support.
Author Contributions
Conceptualization—J.F. and J.D.; Data collection—J.F., J.D., M.G.B.; Writing—original draft, J.F.; Writing, review, editing—J.F., J.D., M.G.B. All authors have read and agreed to the published version of the manuscript.
Funding
This case report received no external funding.
Institutional Review Board Statement
Ethical oversight and IRB approval were not required because the clinical care described in this article was not part of a research project. The results presented in this paper have been deidentified, will not affect the patient’s clinical care, nor will they have the potential to cause the patient any harm. This article is a scholarly report of a clinical cohort of patients with PWS who received pharmacogenetic testing as part of their medical care at one of three specialty programs across the USA. This report will both inform and hopefully improve the quality of care of patients with PWS.
Informed Consent Statement
Patient consent for this case report was waived because all pertinent private and protected health information about each patient was eliminated prior to collation and analysis of data, except for the genetic subtype of PWS and the pharmacogenetic genotype.
Data Availability Statement
The data presented in this paper were obtained through public websites including literature review using PUBMED, public access websites, such as https://www.pharmgkb.org and http://drug-interactions.medicine.iu.edu [32], and standard reference books by Mrazek [38] and the PWS management book edited by Butler, Lee and Whitman [1]. The pharmacogenetic data obtained from the testing of our patients can be found in the Table A2, Table A3 and Table A4. All the tables and figures contained in this article are derived from these data.
Conflicts of Interest
The authors declare no conflict of interest.
Appendix A
genes-12-00152-t0A1_Table A1 Table A1 Chromosome location for genes producing cytochrome P450 (CYP) enzymes, their common polymorphisms and phenotypic activity [38]; https://www.pharmgkb.org.
Cytochrome P450 Chromosome Location Common Alleles and Phenotypic Action
CYP2D6 22q13.1 *1 (wildtype); *2 (normal); *2A (increased action); *9, *10, *17, *41 (decreased action); *4, *3,* 5, *6 (inactive).
CYP2B6 19q13.2 *1 (wildtype); *2, *5, *17 (normal); *4, *22 (increased action); *6, *7, *9, *16, *19, *20 (decreased action); *8, *12, *13, *18,*24 (inactive).
CYP2C19 10q24.1–q24.3 *1 (wildtype); *11, *13, *15, *18,*28 (normal); *17 (increased action); *9, *10, *16, *19, *25, *26 (decreased action); *2-*8, *22, *24, *35-*37 (inactive).
CYP2C9 10q23.33 *1 (wildtype); *9 (normal); *2, *3 (decreased action); *6, *15, *25 (inactive).
CYP3A4 7q22.1 *1 (wildtype); *7, *9, *10 (normal); *1B, *2-*6, *8, *11-*13, *15-*18, *22 (decreased action); *20, *26 (inactive).
CYP1A2 15q24.1 *1, *1A, *1B (wildtype); *1D (T-delT), *1F(C/A) increased action when induced; *1C(G/A); *1K(C/A); *3(T/C), *4(A/T), *7(G/A) (decreased action).
genes-12-00152-t0A2_Table A2 Table A2 Pharmacogenetic data for cytochrome P450 enzymes in PWS cohort with Deletion: Testing laboratory, CYP genotype, and CYP phenotype.
CYP2D6 CYP2B6 CYP2C19 CYP2C9 CYP3A4 CYP1A2
Lab Alleles PT Alleles PT Alleles PT Alleles PT Alleles PT Alleles PT
GS *1/*1 EM *1/*6 IM *1/*1 EM *1/*1 EM *1/*1 EM *1F/*1B UM
GS *4/*41 PM *1/*6 IM *1/*17 EM *1/*1 EM *1/*1 EM *1A/*1B EM
GS *2/*17
DUP EM *1/*6 IM *1/*17 EM *1/*1 EM *1/*1 EM *1A/*1B EM
GS *2/*4 IM *1/*1 EM *1/*17 EM *1/*1 EM *1/*22 IM *1A/*1B EM
GS *1/*10 IM *1/*6 IM *17/*17 UM *1/*1 EM *1/*1 EM *1F/*1B UM
GL *2A/*5 IM *1/*5 EM *1/*1 EM NA NA *1/*22 IM *1A/*1A EM
GS *4/*4 PM *1/*1 EM *1/*1 EM *1/*1 EM *1/*1 EM *1F/*1B UM
GS *1/*41 EM *1/*1 EM *1/*2 IM *1/*1 EM *1/*1 EM *1F/*1B UM
GS *1/*9 IM *1/*6 IM *1/*17 EM *1/*1 EM *1/*1 EM *1F/*1B UM
GS *1/*9 IM *1/*1 EM *1/*17 EM *1/*2 IM *1/*1 EM *1A/*1B EM
GL *1/*4 IM *1/*5 EM *1/*1 EM *1/*2 IM *1/*1 EM *1A/*1F HI
GL *1/*4 IM *1/*6 IM *1/*1 EM *1/*1 EM *1/*1 EM *1C*1F/*1F HI
GS *1/*4 IM *1/*1 EM *1/*1 EM *1/*2 IM *1/*1 EM *1F/*1B UM
GL *1/*1 EM *5/*6 IM *1/*17 UM *1/*1 EM *1/*1 EM *1F/*1F HI
Key: Genesight (GS), Genelex (GL); PT—Phenotype; PM—poor metabolizer; IM—intermediate metabolizer; EM—extensive (normal) metabolizer; UM—ultra-rapid metabolizer; NA—result not available; HI—hyperinducer.
genes-12-00152-t0A3_Table A3 Table A3 Pharmacogenetic data for cytochrome P450 enzymes in PWS cohort with UPD: Testing laboratory, CYP genotype, and CYP phenotype.
CYP2D6 CYP2B6 CYP2C19 CYP2C9 CYP3A4 CYP1A2
Lab Alleles PT Alleles PT Alleles PT Alleles PT Alleles PT Alleles PT
GS *1/*1 EM *1/*6 IM *1/*1 EM *1/*1 EM *1/*1 EM *1C*1D/
*1F*1B UM
GS *1/*4 IM *1/*6 IM *1/*1 EM *1/*3 IM *1/*1 EM *1F/*1B UM
GS *1/*2A EM *1/*6 IM *1/*1 EM *1/*1 EM *1/*1 EM *1F/*1B UM
GS *1/*41 EM *1/*6 IM *1/*17 EM *1/*1 EM *1/*1 EM *1A*1C/
*1B IM
GS *1/*4 EM *1/*6 IM *1/*1 EM *1/*2 IM *1/*1 EM *1D*1E/
*1F*1B UM
GS *2A/*4 EM *1/*1 EM *2/*17 EM *1/*1 EM *1/*1 EM *1F/*1B UM
GS *1/*2A EM *1/*6 IM *1/*17 EM *1/*1 EM *1/*1 EM *1F/*1B UM
GS *10/*41 PM *1/*1 EM *1/*17 EM *1/*1 EM *1/*1 EM *1A/*1B EM
GS *1/*5 IM *1/*1 EM *1/*1 EM *1/*2 IM *1/*1 EM *1F/*1B UM
GS *1/*3 IM NA NA *1/*1 EM *1/*3 IM NA NA *1F/*1B UM
GS *1/*41 EM *1/*6 IM *2/*17 EM *1/*1 EM *1/*1 EM *1A/*1A EM
GS *4/*41 PM *1/*1 EM *1/*17 EM *1/*1 EM *1/*1 EM *1A/*1A EM
GS *4/*41 PM *1/*6 IM *1/*1 EM *2/*2 PM *1/*1 EM *1F/*1B UM
GM *2/*41
Dup EM *1/*6 EM *1/*1 EM *1/*1 EM *1/*22 IM *1A/*1F EM
GS *1/*4 IM *1/*1 EM *1/*1 EM *1/*2 IM *1/*22 IM *1F/*1B UM
GL *1/*2A EM *2/*5 IM *1/*1 EM *1/*2 IM *1/*1B IM *1F/*1F HI
Key: Genesight (GS), Genelex (GL), and Genomind (GM); PT—Phenotype; PM—poor metabolizer; IM—intermediate metabolizer; EM—extensive (normal) metabolizer; UM—ultra-rapid metabolizer; HI—hyperinducer; NA—result not available.
genes-12-00152-t0A4_Table A4 Table A4 Pharmacogenetic data for cytochrome P450 enzymes in PWS Unspecified cohort (DNA methylation positive, subtype unspecified): Testing laboratory, CYP genotype, and CYP phenotype.
CYP2D6 CYP2B6 CYP2C19 CYP2C9 CYP3A4 CYP1A2
Lab Alleles PT Alleles PT Alleles PT Alleles PT Alleles PT Alleles PT
GS *2A/*4 EM *1/*1 EM *1/*1 EM *1/*3 IM *1/*1 EM *1A/*1B EM
GS *1/*1 EM *1/*1 EM *1/*8 IM *1/*2 IM *1/*1 EM *1F/*1B UM
GS *1/*1 EM *1/*1 EM *1/*1 EM *1/*2 IM *1/*1 EM *1A/*1B EM
GS *6/*41 PM *1/*1 EM *1/*17 EM *1/*1 EM *1/*1 EM *1F/*1B UM
GS *1/*1 EM *1/*1 EM *1/*2 IM *1/*1 EM *1/*1 EM *1A/*1A EM
Key: Genesight (GS); PT—Phenotype; PM—poor metabolizer; IM—intermediate metabolizer; EM—extensive (normal) metabolizer; UM—ultra-rapid metabolizer.
genes-12-00152-t0A5_Table A5 Table A5 Ratios of Cytochrome P450 phenotypes among PWS genetic subtypes.
PWS CYP2D6 CYP2B6 CYP2C19 CYP2C9 CYP3A4 CYP1A2
DEL 4EM:8IM:2PM 7EM:7IM 2UM:11EM:1IM 10EM:3IM 12EM:2IM 6UM:3HI:5EM
UPD 9EM:4IM:3PM 6EM:9IM 16EM 9EM:6IM:1PM 12EM:3IM 10UM:1HI:4EM:1IM
UnSpec 4EM:1PM 5EM 3EM:2IM 2EM:3IM 5EM 2UM:3EM
Key: UM—ultra-rapid metabolizer; EM—extensive (normal) metabolizer; IM—intermediate metabolizer; PM—poor metabolizer; HI—hyperinducer; DEL—deletion; UPD—uniparental disomy; UnSpec—unspecified genetic subtype.
Figure 1 Distribution of Cytochrome P450 phenotypes among the combined PWS cohort (n = 35) compared to predicted, normative populations [https://www.pharmgkb.org/page/cyp2d6RefMaterials]; [https://www.pharmgkb.org/page/cyp2b6RefMaterials]; [https://www.pharmgkb.org/page/cyp2c19RefMaterials]; [https://www.pharmgkb.org/page/cyp2c9RefMaterials]; [20,27]. Key: UM—ultra-rapid metabolizer; EM—extensive metabolizer; IM—intermediate metabolizer; PM—poor metabolizer.
Figure 2 Distribution of Cytochrome P450 phenotypes among PWS DEL (n = 14) compared to normative populations [https://www.pharmgkb.org/page/cyp2d6RefMaterials]; [https://www.pharmgkb.org/page/cyp2b6RefMaterials]; [https://www.pharmgkb.org/page/cyp2c19RefMaterials]; [https://www.pharmgkb.org/page/cyp2c9RefMaterials]; [20,27]. Key: UM—ultra-rapid metabolizer; EM—extensive metabolizer; IM—intermediate metabolizer; PM—poor metabolizer.
Figure 3 Distribution of Cytochrome P450 phenotypes among the PWS UPD cohort (n = 16) compared to normative populations [https://www.pharmgkb.org/page/cyp2d6RefMaterials]; [https://www.pharmgkb.org/page/cyp2b6RefMaterials]; [https://www.pharmgkb.org/page/cyp2c19RefMaterials]; [https://www.pharmgkb.org/page/cyp2c9RefMaterials]; [20,27]. Note: There were 15 results for cytochromes 2B6 and 3A4. Key: UM—ultra-rapid metabolizer; EM—extensive metabolizer; IM—intermediate metabolizer; PM—poor metabolizer.
Figure 4 Distribution of Cytochrome P450 phenotypes among PWS Unspecified genetic subtype, (n = 5) compared to normative populations [https://www.pharmgkb.org/page/cyp2d6RefMaterials]; [https://www.pharmgkb.org/page/cyp2b6RefMaterials]; [https://www.pharmgkb.org/page/cyp2c19RefMaterials]; [https://www.pharmgkb.org/page/cyp2c9RefMaterials]; [20,27]. Key: UM—ultra-rapid metabolizer; EM—extensive metabolizer; IM—intermediate metabolizer; PM—poor metabolizer.
Figure 5 Cytochrome 2D6 allele frequencies among a referred cohort of patients with PWS DEL (n = 14), UPD (n = 16) and combined cohort (ALL PWS, n = 35) compared to predicted, American normative data [http://www.pharmgkb.org/page/cyp2d6RefMaterials].
Figure 6 Cytochrome 2D6 diplotype frequencies among PWS DEL (n = 14), UPD (n = 16) and combined cohort (ALL PWS, n = 35) compared to predicted, American normative data [http://www.pharmgkb.org/page/cyp2d6RefMaterials].
Figure 7 Cytochrome 2B6 allele frequency and diplotype distribution in a referred cohort with PWS DEL (n = 14), UPD (n = 15) and combined cohort (ALL PWS, n = 34) compared to the predicted, European normative data [http://www.pharmgkb.org/page/cyp2b6RefMaterials].
Figure 8 CYP2C19 allele frequency and distribution of diplotypes in PWS DEL (n = 14), UPD (n = 16) and combined cohort (ALL PWS, n = 35) compared to the predicted normative American data [http://www.pharmgkb.org/page/cyp2c19RefMaterials].
Figure 9 CYP2C9 allele and diplotype frequencies in PWS DEL (n = 13), UPD (n = 16) and combined cohort (ALL PWS, n = 34) compared to American normative, predicted data [http://www.pharmgkb.org/page/cyp2c9RefMaterials].
Figure 10 CYP3A4 allele and diplotype frequencies among PWS DEL (n = 14), UPD (n = 15), and combined cohort (ALL PWS, n = 34) compared to Caucasian normative, predicted data [33,34].
Figure 11 CYP1A2 allelic frequencies in PWS DEL (n = 14), UPD (n = 16) and combined cohort (ALL PWS, n = 35) compared to normative Caucasian predicted values [37,38].
genes-12-00152-t001_Table 1 Table 1 Frequencies of CYP phenotypes in PWS cohort compared to a typical population, and substrates most likely to be affected based on drug utilization data from the NIH PWS Registry.
CYP Gene/Metabolizer Phenotype CYP Phenotype Frequency Substrate Frequency
PWS Referred Cohort Typical Population PWS Clinic Patients
DEL UPD Unspec All PWS Frequency/Reference NIH PWS Registry [12,13]
CYP2D6 (n = 14) (n = 16) (n = 5) (n = 35) (n = 56,945) (n = 265)
EM 28.5% † 56.3% 4 48.6% 63.6% http://www.pharmgkb.org/page/cyp2d6RefMaterials Fluoxetine (21.9%), Risperidone (14%), Sertraline (14%), Aripiprazole (9.8%), Citalopram (8.7%), Escitalopram (5.7%), Paroxetine (4.5%), Bupropion (4.5%), Amphetamine (4.2%), Clonidine (3%), Ziprasidone (3%)
IM 57.1% * 25.0% 0 34.0% 23.6%
PM 14.3% † 18.8% * 1 17.1% * 2.2%
CYP2B6 (n = 14) (n = 15) (n = 5) (n = 34) (n = 56,945)
EM 50% 40% 5 53% 43% http://www.pharmgkb.org/page/cyp2B6RefMaterials Sertraline (14%), Bupropion (4.5%)
IM 50% 60% 0 47% 39%
CYP2C19 (n = 14) (n = 16) (n = 5) (n = 35) (n = 56,945)
EM 78.6% 100% † 3 85.7% 76.4% http://www.pharmgkb.org/page/cyp2C19RefMaterials Fluoxetine (21.9%), Sertraline (14%),
Citalopram (8.7%), Escitalopram (5.7%)
IM 7.1% 0% † 2 8.6% 21.4%
UM 14.3% † 0% 0 5.7% 0.74%
CYP2C9 (n = 14) (n = 16) (n = 5) (n = 35) (n = 56,945)
EM 76.9% 56.2% * 2 61.8% * 83.2% http://www.pharmgkb.org/page/cyp2C9RefMaterials Fluoxetine (21.9%), Sertraline (14%),
Valproate (6.4%)
IM 23.0% 37.5% * 3 35.3% * 16.4%
PM 0% 6.3% † 0 2.9% 0%
CYP3A4 (n = 14) (n = 15) (n = 5) (n = 34) (n = 5789)
EM 85.7% † 80.0% † 5 85.3% * 97.3% Zhou et al., 2017 [20] Risperidone (14%), Sertraline (14%),
Modafinil (12.8%), Aripiprazole (9.8%),
Citalopram (8.7%), Clonazepam (6.8%),
Escitalopram (5.7%), Bupropion (4.5%),
Ziprasidone (3%)
IM 14.3% † 20.0% † 0 14.7% * 2.7%
CYP1A2 (n = 14) (n = 16) (n = 5) (n = 35) (n = 183)
UM 42.9% * 62.5% * 2 51.4% * 0% Muscat et al., 2008 [27] Fluvoxamine (1.1%), Haloperidol (1.1%),
Thioridazine (1.1%), Olanzapine (0.4%),
Chlorpromazine (0.4%), Imipramine (0.4%)
EM/HI 57.1% 31.2% 3 45.7% 37.0%
IM 0% † 6.25% † 0 2.85% 54.0%
PM 0% 0% 0 0% † 10.0%
KEY: Phenotype% = within group comparison; (*) = statistical significance by chi-square, p < 0.05; (†) = statistical significance by chi-square, p < 0.05, but results may not be reliable due to small cell size. The frequency of psychotropic medications in this table is derived from the NIH PWS Registry [12,13]. Only the most frequently prescribed medications are listed for each cytochrome. This data reflects regional prescribing practices; it does not reflect treatment efficacy, nor does it constitute recommended treatment.
genes-12-00152-t002_Table 2 Table 2 Psychotropic medications, nutraceuticals, agents, and related cytochrome P450 enzymes.
Cytochrome CYP1A2 CYP2B6 CYP2C19 CYP2C9 CYP2D6 CYP3A4-5
Substrate Acetaminophen Amitriptyline, 3 Benztropine, 2
Caffeine
Cannabidiol, 3
Chlorpromazine, 2 Clomipramine, 2 Clozapine, 1
Duloxetine, 1
Estradiol
Fluvoxamine, 1 Haloperidol, 3 Imipramine, 2
Melatonin
Mirtazapine, 3 Olanzapine, 1
Pimozide
Propranolol, 1
Tacrine
Thioridazine, 2 Bupropion, 1 Selegiline Sertraline, 1 Sibutramine Amitriptyline, 1 Benztropine, 3 Cannabis, 1 Cannabidiol, 2 Citalopram, 1 Clomipramine, 1
Clozapine, 2 Diazepam Doxepin, 2 Escitalopram, 1 Estradiol
Fluoxetine,3
Imipramine, 1 Nortriptyline, 2
Omeprazole
Phenytoin
Progesterone, 1
Sertraline, 2
Testosterone, 1
Venlafaxine, 2 Amitriptyline, 2
Benztropine, 4
Cannabidiol, 3
Cannabis, 3
Celecoxib
Fluoxetine, 2
Ibuprofen, 1
Progesterone, 2
Sertraline, 3
Valproate
Warfarin Amitriptyline, 2
Amphetamine
Aripiprazole, 2
Asenapine
Atomoxetine
Bupropion, 2
Cannabidiol, 3
Chlorpromazine, 1
Citalopram, 3
Clomipramine, 4
Clonidine
Clozapine, 3
Desipramine, 4
Dextromethorphan
Diphenhydramine
Doxepin, 1
Escitalopram, 3
Fluoxetine, 1
Fluvoxamine, 2
Haloperidol, 1
Hydroxyzine
Imipramine, 2
Mirtazapine, 2
Nortriptyline, 1
Olanzapine, 2
Paroxetine, 1
Perphenazine
Propranolol, 2
Quetiapine, 2
Ranitidine
Risperidone, 1
Sertraline, 2
Thioridazine, 1
Trazadone, 2
Venlafaxine, 1
Ziprasidone, 3 Alprazolam
Aripiprazole, 1
Benztropine, 1
Bupropion, 2
Buspirone
Cannabis, 2
Cannabidiol, 1
Cisapride
Citalopram, 2
Clomipramine, 3
Clonazepam
Carbamazepine
Desvenlafaxine
Dextromethorphan
Duloxetine, 3
Escitalopram, 2
Estradiol
Erythromycin
Fexofenadine
Guanfacine
Haloperidol, 2
Loratadine
Lurasidone
Mirtazapine, 1
Modafinil
Pimozide
Progesterone, 3
Quetiapine, 1
Risperidone, 2
Sertraline, 3
Testosterone
Trazadone, 1
Tiagabine
Ziprasidone, 1
Zolpidem
Inhibitor Cannabidiol
Cannabis
Celecoxib
Cimetidine
Citalopram
Ciprofloxacin
Clarithromycin
Erythromycin
Estradiol
Fluvoxamine
Isoniazid
Ketoconazole
Modafinil Cannabidiol
Cannabis Cannabidiol
Cannabis
Cimetidine
Contraceptives
Fluconazole
Fluoxetine
Fluvoxamine
Indomethacin
Isoniazid
Ketoconazole
Lansoprazole
Modafinil
Omeprazole
Oxcarbazepine
Probenecid
Topiramate Cannabidiol
Cannabis
Cimetidine
Contraceptives
Fluconazole
Fluoxetine
Fluvoxamine
Isoniazid
Ketoconazole
Methylphenidate
Modafinil
Omeprazole
Paroxetine
Sertraline
Sulfonamides
Tacrine Asenapine
Bupropion
Cannabis
Cannabidiol
Diphenhydramine
Fluoxetine
Goldenseal
Haloperidol
Hydroxyzine
Methylphenidate
Paroxetine
Propranolol
Quinidine
Ranitidine
Cannabidiol
Cimetidine
Ciprofloxacin
Clarithromycin
Cyclosporine
Erythromycin
Goldenseal
Grapefruit juice
Isoniazid
Ketoconazole
Prednisone
Sertraline
Verapamil
Inducer Carbamazepine
Cruciferous vegetables
Cannabis (smoke)
Char-grilling
Omeprazole
Phenytoin
St. John’s wort Tobacco (smoke)
Carbamazepine
Modafinil
Phenytoin
Phenobarbital
Rifampin Rifampin Carbamazepine
Phenobarbital
Rifampin
St. John’s wort Carbamazepine
Cruciferous vegetables
Ginseng
Modafinil
Phenytoin
Rifampin
St. John’s wort
Note: Pertinent information about medications and cytochrome selectivity in Table 2 was obtained from multiple sources including literature review, [http://www.pharmgkb.org/pathways] and the Flockhart website [32]. Substrates metabolized by more than one cytochrome are numbered 1, 2, or 3 to indicate binding affinity and to delineate primary, secondary, or tertiary pathways. Inhibition in one pathway has the potential to shunt metabolism through another pathway.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. | ARIPIPRAZOLE | DrugsGivenReaction | CC BY | 33498922 | 19,918,290 | 2021-01-24 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Drug ineffective'. | Angiogenesis Inhibitors as Anti-Cancer Therapy Following Renal Transplantation: A Case Report and Review of the Literature.
Solid organ transplant recipients on long-term immunosuppressive medication are at increased risk of developing malignancy, and treatment of advanced cancers with angiogenesis inhibitors in this context has not been widely studied. We present a case of recurrent high-grade serous ovarian carcinoma treated with paclitaxel and bevacizumab in the context of prior renal transplantation where the patient responded well to treatment with controlled toxicities, discussing the potential for increased rates of adverse events and drug interactions in this select population.
1. Introduction
Angiogenesis inhibitors such as vascular endothelial growth factor (VEGF) monoclonal antibodies and tyrosine kinase inhibitors (Figure 1) are standard treatments across various cancer subtypes. In advanced high-grade serous ovarian cancer (HGSOC), bevacizumab is part of the standard of care as maintenance therapy in front-line and recurrent disease [1]. Although the toxicity profile is generally well-managed, nephrotoxicity manifesting as proteinuria remains an important adverse event that requires close monitoring [2]. There is limited literature surrounding angiogenesis inhibitors as anti-cancer treatment in patients who have received solid organ transplants, and thus this report presents a case demonstrating safety of angiogenesis inhibition as anti-cancer therapy in a patient with a stable renal transplant.
2. Case Description
Consent: Fully informed, voluntary, written consent has been obtained to include patient information and publish this report.
Ethics: As this is a case report with fewer than three patients, institutional approval was not sought as per the University Health Network Research Ethics Board guidance document on case reports.
A 47-year-old woman presented with several months of dyspnea and abdominal distension to Princess Margaret Cancer Centre in February 2019 (Figure 2). CT revealed a 13.9 cm pelvic mass with peritoneal carcinomatosis and ascites. Omental biopsy confirmed HGSOC, and she received neoadjuvant platinum-based chemotherapy with excellent tolerance and no renal complications before proceeding to interval debulking in June 2019. There was no visible residual disease, and diagnosis of HGSOC was confirmed, germline and somatic BRCA wild type. She completed three cycles of adjuvant chemotherapy.
Her background was significant for IgA nephropathy, which resulted in progressive chronic kidney disease for 18 years prior to living donor kidney transplant in 2016. Both she and her donor were cytomegalovirus-positive, and she developed cytomegalovirus-associated colitis shortly post-transplantation. Her initial immunosuppression consisted of basiliximab induction followed by tacrolimus, mycophenolic acid, and steroids. She developed antibody-mediated rejection one week post-transplant, which was treated with plasmapheresis, immunoglobulin, and an increase in steroid dose. This was repeated three months later due to biopsy confirming ongoing antibody-mediated rejection. Mycophenolic acid was stopped upon HGSOC diagnosis. She has remained medication-adherent, with regular serum tacrolimus levels within the target range (most recently, 5.3 micrograms/L) and no signs of chronic graft rejection.
Her other comorbidities include diet-controlled, steroid-induced diabetes mellitus; ductal breast carcinoma in situ requiring wide local excision in 2011; asthma; reflux disease; hypertension; and hyperlipidemia. Other medications include prednisone 5 mg daily, acetylsalicylic acid, bisoprolol, trimethoprim–sulfamethoxazole, vitamin D, and inhaled salbutamol as needed.
In December 2019, she developed recurrence in the peritoneum and retroperitoneal lymph nodes, signifying platinum resistance. In January 2020, she commenced weekly paclitaxel 80 mg/m2 with bevacizumab 10 mg/kg every two weeks, and at that time, tacrolimus dose was reduced to 2 mg daily to aim for a serum level of 5 micrograms/L. She continued this therapy for over 6 months, and continued on the same dose of immunosuppression throughout with tacrolimus levels ranging between 3 and 7.6 micrograms/L. Serial imaging and Ca125 confirmed good response to treatment with reduction in size of tumor deposits. Her albumin/creatinine ratio was normal at 0.9 prior to the diagnosis of ovarian cancer, and most recently has been 15.6, signifying microalbuminuria; this has been monitored via urinalysis, which has consistently reported protein as negative or trace. Her most recent estimated glomerular filtration rate was 55 mL/min, similar to pre-diagnosis, and creatinine levels have mostly fluctuated between 95 and 120 umol/L.
Her course has been complicated by grade 2 hypertension (up to 145/95 mmHg) and non-cardiac chest pain, for which amlodipine was switched to ramipril 10 mg daily, and bevacizumab was withheld on two different occasions. Furthermore, she developed a brief period of Kidney Disease: Improving Global Outcomes (KDIGO) stage 1 acute kidney injury (AKI) of pre-renal etiology in September 2020 with creatinine of 160 umol/L, which self-resolved following withdrawal of bevacizumab, and was resumed after a 1-month-long break with no further episodes of kidney injury.
3. Discussion
Nephrotoxicity with angiogenesis inhibitors is relatively common, with proteinuria occurring in over 60% of patients [2]. Most cases are low-grade, transient, and do not require interventions or dose delays; however, more persistent, severe cases presenting as AKI and nephrotic syndrome can occur [2,3]. Risk factors associated with high-grade proteinuria include increased dose, prolonged administration, pre-existing renal disease, and administration of concurrent chemotherapy [4,5].
The pathophysiology of VEGF inhibitor-induced proteinuria remains unclear. Within a normal kidney, VEGF is produced by podocytes, and VEGF receptors are typically present on the glomerular and peritubular endothelium in addition to mesangial cells [6]. Inhibition of VEGF is thought to cause loss of endothelial fenestrations, podocyte injury and reduce endothelial proliferation, ultimately causing disruption of glomerular membranes [6]. Some cases have also demonstrated subacute thrombotic microangiopathy with endotheliosis and membranoproliferative changes [7]. Another manifestation of nephrotoxicity that is commonly seen is hypertension, occurring in more than a third of patients, which arises due to various mechanisms of renal vascular injury including inhibition of nitric oxide, rarefaction of microvasculature, and neuroendocrine dysregulation [8,9]. It is also hypothesized to increase intraglomerular pressure and ultrafiltration, leading to proteinuria [10].
The lack of clarity surrounding pathophysiology of VEGF inhibitor-induced nephrotoxicity [11] is demonstrated by the heterogeneity of published reports on glomerulopathy and other manifestations, including minimal change disease, collapsing glomerulopathy, membranoproliferative glomerulonephritis, focal segmental glomerulosclerosis, cryoglobulinemic glomerulonephritis, acute tubular necrosis, and interstitial nephritis [7]. Furthermore, worsening kidney disease can further exacerbate hypertension, which may perpetuate AKI [10].
Treatment for low-grade proteinuria usually includes an angiotensin-converting enzyme (ACE) inhibitor or an angiotensin receptor blocker to reduce glomerular filtration pressure, and anti-VEGF treatment can be continued providing stable proteinuria. However, proteinuria may worsen to the nephrotic range (>3 g in 24 h) with nephrotic syndrome. Even after cessation of VEGF inhibitor therapy, there are documented cases of persistent proteinuria [12].
3.1. Angiogenesis Inhibitors Post-Transplant
The role of the VEGF pathway in the pathogenesis of post-transplant complications is poorly understood, with preliminary reports across various organ transplants showing hypothesis-generating results [13,14,15,16]. Upregulation of angiogenesis factors was associated with increased allograft vasculopathy, bronchiolitis obliterans, and recurrence of hepatocellular carcinoma in cardiac, pulmonary, and liver transplants, respectively [13,15,16,17]. In renal transplants, VEGF is thought to be upregulated in acute and chronic rejection, particularly associated with cyclosporine [14].
As the first VEGF inhibitor used for anti-cancer treatment, bevacizumab was approved by the United States Food and Drug Administration in 2004, and is now licensed for use in many cancers [18] (Table 1). In ovarian cancer, patients who were prior recipients of solid organ transplants or who were receiving immunosuppressive therapies were not excluded from randomized bevacizumab trials, but those with pre-existing uncontrolled hypertension or renal dysfunction based on serum creatinine ≥ 1.6 mg/dL or proteinuria > 1 g per 24 h were excluded [19,20,21]. Similarly, in other large randomized angiogenesis inhibitor studies across other tumor sites, prior solid organ transplant or use of immunosuppressants is not an exclusion criterion, aside from studies involving immune checkpoint inhibitors [22]. Identification of patients enrolled in large angiogenesis inhibitor trials who had received prior transplants could potentially make for an interesting post-hoc pooled analysis.
Similarly, reports on angiogenesis inhibition in solid organ transplant patients remain scant in the literature, as highlighted by a review on bevacizumab toxicity by Fenoglio et al [9]. Musri et al. reported a case of colorectal cancer post-renal transplantation with baseline proteinuria, which significantly worsened on administration of intravenous 5-fluorouracil, irinotecan, oxaliplatin, and bevacizumab [23]. Cheungpasitporn et al. described two cases with renal allograft dysfunction following administration of intravitreal bevacizumab, aflibercept, or ranibizumab [24]. Doses of anti-angiogenics were lower but not specified within this report. Although neither case proved causality with anti-VEGF therapy, one was diagnosed with phospholipase A2 receptor-negative membranous nephropathy, and the second revealed acute and chronic antibody-mediated rejection with glomerular thrombi and transplant glomerulopathy. Jonkers and Buren reported a case of worsening IgA nephropathy presenting with nephrotic-range proteinuria post-renal transplantation on sorafenib [25]. These reports highlight the potential severe nephrotoxicity known to be associated with angiogenesis inhibitors; however, there remain few documented positive experiences with angiogenesis inhibitor use in the post-transplant setting. Given the prevalence of nephrotoxicity with these agents, these considerations are particularly prudent for renal transplant recipients, but reports in other organ transplants remain similarly scarce.
3.2. Medication Interactions
Solid organ transplant recipients frequently take maintenance immunosuppressive agents, including but not limited to corticosteroids, calcineurin inhibitors, anti-proliferative agents, and mTOR inhibitors, which are associated with various complications and drug interactions (Table 2).
In the case presented, tacrolimus levels were measured every few months to be within the therapeutic range; this is significant as potential drug interactions between transplant medications and angiogenesis inhibitors involve pharmacokinetic and pharmacodynamic interactions. Pharmacokinetic interactions typically occur due to cytochrome P450 enzyme (CYP) and P-glycoprotein (P-gp) drug transport systems [26,27,28], and risk of competitive metabolism as substrates for the same enzyme or transporter may increase serum levels. Axitinib and sorafenib are CYP3A4 and P-gp substrates, and cabozantinib and pazopanib are substrates and inhibitors of both enzyme systems [29,30,31,32]. Drug interactions are well-described for CYP3A4 substrates cyclosporine, tacrolimus, and sirolimus, with metabolic inhibition leading to increased immunosuppressant concentrations (~20%) [28]. Pharmacodynamic interactions primarily concern cumulative toxicities between these two classes (Table 2) [26,27,33,34]. Interestingly, Onodera et al. reported upon a case of metastatic colorectal cancer post-renal transplant where a patient was administered five cycles of 5-fluorouracil, oxaliplatin, and bevacizumab where although severe proteinuria occurred, serum tacrolimus levels were not affected throughout the course of treatment [35]. This report remains one of the only cases in the literature that demonstrates stability of immunosuppression whilst on bevacizumab post-transplant.
3.3. Long-Term Adverse Events
One of the leading causes of morbidity and mortality in solid organ transplant recipients is malignancy, most commonly non-melanomatous skin cancers [36]. Other malignancies such as colorectal, kidney, and cervical cancers are also prevalent in the post-transplant context, and angiogenesis inhibitors such as bevacizumab are commonly used in metastatic disease [18] (Table 1). Surveillance recommendations within transplant recipients are variable across the globe due to a paucity of robust screening trials [37]. Other long-term complications associated with organ transplantation and prolonged immunosuppressant use include cardiovascular disease, diabetes mellitus, hypertension, and infection associated with cytopenia. In the patient presented, adverse events have not outweighed benefits of ongoing treatment, but this will need close monitoring given the risk of overlapping toxicities as long-term adverse event data remain limited [3,38].
4. Conclusions
Whilst there are minimal data justifying that bevacizumab or other angiogenesis inhibitors are unsafe in the post-transplant setting, there is similarly scarce literature demonstrating safe administration, as in the patient presented. As life expectancy continues to improve with increasing indications for transplantation, long-term risks for malignancy with prolonged immunosuppression are increasingly relevant as a cause of mortality in solid organ transplant recipients. In patients who have undergone renal transplantation, careful consideration of treatment options with risk of nephrotoxicity and close monitoring remains paramount.
Although treating oncologists should remain vigilant about potential drug interactions and overlapping toxicities, these are not necessarily contraindications for agents such as bevacizumab. Treatment decisions should consider the best available evidence, and collating information about toxicity and tolerance from randomized trials and post-approval Phase IV studies would provide detailed information from at-risk subgroups. This calls for a stratified, inclusive approach to allow enrolment of those with chronic diseases and comorbidities in prospective trials, allowing objective assessment of the risk–benefit ratio.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Author Contributions
Conceptualization, L.K., A.M.O.; methodology, L.K., J.D.; software, L.K.; validation, L.K., J.D., J.S., A.K.; resources, L.K.; data curation, L.K., S.L., J.S.; writing—original draft preparation, L.K., J.D.; writing—review and editing, L.K., J.D., K.K., A.K., J.S., S.L., A.M.O.; visualization, L.K., K.K.; supervision, S.L., A.M.O. All authors have read and agreed to the published version of the manuscript.
Funding
This research received no external funding.
Institutional Review Board Statement
As this was a report with fewer than three patients, institutional approval was not sought as per the University Health Network Research Ethics Board guidance document on case reports.
Informed Consent Statement
Informed consent was obtained from all subjects involved in the study and written informed consent has been obtained from the patient to publish this paper.
Data Availability Statement
Not applicable.
Conflicts of Interest
We have read and understood Current Oncology’s policy on disclosing conflicts of interest and declare that we have none.
Abbreviations
VEGFR vascular endothelial growth factor receptor
EGFR epidermal growth factor receptor
FGFR fibroblast growth factor receptor
PDGFR platelet-derived growth factor receptor
SCF stem cell factor
PI3K phosphoinositide 3-kinase
AKT protein kinase B
mTOR mammalian target of rapamycin
RET RET proto-oncogene
AXL AXL receptor tyrosine kinase
TAH–BSO total abdominal hysterectomy and bilateral salpingo-oophorectomy
Ca125 cancer antigen 125
U/ml units per milliliter
AKI acute kidney injury
CYP cytochrome P450
HGSOC high-grade serous ovarian carcinoma
FOLFOX4 5-fluorouracil, leucovorin, and oxaliplatin
IFL irinotecan, leucovorin, and 5-fluorouracil
P-gp P-glycoprotein
Figure 1 Angiogenesis pathways in malignancy and targeted therapies commonly used. Monoclonal antibodies including bevacizumab and ramucirumab and tyrosine kinase inhibitors including sunitinib, sorafenib, lenvatinib, pazopanib, axitinib, cediranib, and cabozantinib have been included (this list is not exhaustive). Receptors including VEGFR, c-kit, epidermal growth factor receptor (EGFR), fibroblast growth factor receptor (FGFR), and PDGFR are displayed here, but certain medications also target other pathways not displayed here such as AXL and RET. The VEGFR pathway intersects with multiple cell signaling pathways, including the PI3K/AKT/mTOR pathway. Created with Biorender.com.
Figure 2 (a) Summary oncology treatment timeline. (b) Ca125 trend from diagnosis to current.
curroncol-28-00064-t001_Table 1Table 1 Food and Drug Administration approved indications for bevacizumab [18].
Cancer Stage Usage
Colorectal Metastatic, first-line 5 mg/kg every two weeks with bolus IFL
10 mg/kg every two weeks with FOLFOX4
Metastatic, recurrent after first-line bevacizumab-containing regimen 5 mg/kg every two weeks, or 7.5 mg/kg every three weeks with fluoropyrimidine–irinotecan, or fluoropyrimidine–oxaliplatin-based chemotherapy
Non-squamous, non-small-cell lung Unresectable, locally advanced, recurrent, or metastatic 15 mg/kg every three weeks with carboplatin and paclitaxel
Glioblastoma Recurrent 10 mg/kg every two weeks
Renal cell Metastatic 10 mg/kg every two weeks with interferon-alfa
Cervical Persistent, recurrent, or metastatic 15 mg/kg every three weeks with paclitaxel and cisplatin, or paclitaxel and topotecan
Epithelial ovarian, fallopian tube, or primary peritoneal III or IV, following surgical resection 15 mg/kg every three weeks with carboplatin and paclitaxel for up to six cycles, followed by 15 mg/kg every three weeks as a single agent for up to 22 cycles
Recurrent, platinum-sensitive 15 mg/kg every three weeks with carboplatin and either paclitaxel (6–8 cycles) or gemcitabine (6–10 cycles) followed by 15 mg/kg every 3 weeks as a single agent
Recurrent, platinum-resistant 10 mg/kg every two weeks with paclitaxel, pegylated liposomal doxorubicin, or topotecan given weekly
15 mg/kg every three weeks with topotecan every three weeks
Hepatocellular Unresectable or metastatic, first-line 15 mg/kg with atezolizumab every three weeks
Abbreviations: mg/kg = milligrams per kilogram; IFL = infusional fluoropyrimidine; FOLFOX4 = 5-fluorouracil, folic acid, and oxaliplatin.
curroncol-28-00064-t002_Table 2Table 2 Drug–drug interactions between post-transplant immunosuppressive medications and angiogenesis inhibitors.
Transplant Medication Potential Interactions with Anti-Angiogenesis Agents [26,27,28,29,30,31,32,33,34]
Cyclosporine
Increased cyclosporine levels and subsequent toxicity due to CYP3A4 and P-gp-mediated drug interactions (e.g., cabozatinib, axitinib, pazopanib, sorafenib, sunitinib)
Tacrolimus
Increased tacrolimus levels and subsequent toxicity due to inhibition of or competition with CYP3A4 metabolism and P-gp-mediated transport (e.g., cabozatinib, axitinib, pazopanib, sorafenib, sunitinib)
Additive impairment of the renal function (e.g., cediranib, axitinib, pazopanib)
Increased risk of QT prolongation with other agents that prolong the QT interval (e.g., cabozantinib, pazopanib, sorafenib, sunitinib)
Mycophenolate mofetil
Exaggerated leukopenia (e.g., ramucirumab, bevacizumab, sunitinib)
Azathioprine
Exaggerated leukopenia (e.g., ramucirumab, bevacizumab, sunitinib)
Sirolimus
Increased tacrolimus levels and subsequent toxicity due to inhibition of or competition with CYP3A4 metabolism and P-gp-mediated transport (e.g., cabozatinib, axitinib, pazopanib, sorafenib, sunitinib)
Additive impairment of the renal function (e.g., cediranib, axitinib, pazopanib)
Additive impairment of wound healing
Everolimus
Increased everolimus levels and subsequent toxicity due to inhibition of CYP3A4 metabolism and P-gp-mediated transport (e.g., cabozantinib, pazopanib)
Additive impairment of the renal function (e.g., cediranib, axitinib, pazopanib)
Additive impairment of wound healing
Corticosteroids
Competitive CYP3A4 metabolism (e.g., prednisone) with other CYP3A4 substrates (e.g., cabozantinib, axitinib, pazopanib, sorafenib, sunitinib)
Abbreviations: CYP3A4 = cytochrome P450 3A4; P-gp = P-glycoprotein. | ALBUTEROL, ASPIRIN, BASILIXIMAB, BEVACIZUMAB, BISOPROLOL, MYCOPHENOLIC ACID, PACLITAXEL, PREDNISONE, SULFAMETHOXAZOLE\TRIMETHOPRIM, TACROLIMUS, VITAMIN D NOS | DrugsGivenReaction | CC BY | 33499164 | 19,169,975 | 2021-01-22 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Off label use'. | Angiogenesis Inhibitors as Anti-Cancer Therapy Following Renal Transplantation: A Case Report and Review of the Literature.
Solid organ transplant recipients on long-term immunosuppressive medication are at increased risk of developing malignancy, and treatment of advanced cancers with angiogenesis inhibitors in this context has not been widely studied. We present a case of recurrent high-grade serous ovarian carcinoma treated with paclitaxel and bevacizumab in the context of prior renal transplantation where the patient responded well to treatment with controlled toxicities, discussing the potential for increased rates of adverse events and drug interactions in this select population.
1. Introduction
Angiogenesis inhibitors such as vascular endothelial growth factor (VEGF) monoclonal antibodies and tyrosine kinase inhibitors (Figure 1) are standard treatments across various cancer subtypes. In advanced high-grade serous ovarian cancer (HGSOC), bevacizumab is part of the standard of care as maintenance therapy in front-line and recurrent disease [1]. Although the toxicity profile is generally well-managed, nephrotoxicity manifesting as proteinuria remains an important adverse event that requires close monitoring [2]. There is limited literature surrounding angiogenesis inhibitors as anti-cancer treatment in patients who have received solid organ transplants, and thus this report presents a case demonstrating safety of angiogenesis inhibition as anti-cancer therapy in a patient with a stable renal transplant.
2. Case Description
Consent: Fully informed, voluntary, written consent has been obtained to include patient information and publish this report.
Ethics: As this is a case report with fewer than three patients, institutional approval was not sought as per the University Health Network Research Ethics Board guidance document on case reports.
A 47-year-old woman presented with several months of dyspnea and abdominal distension to Princess Margaret Cancer Centre in February 2019 (Figure 2). CT revealed a 13.9 cm pelvic mass with peritoneal carcinomatosis and ascites. Omental biopsy confirmed HGSOC, and she received neoadjuvant platinum-based chemotherapy with excellent tolerance and no renal complications before proceeding to interval debulking in June 2019. There was no visible residual disease, and diagnosis of HGSOC was confirmed, germline and somatic BRCA wild type. She completed three cycles of adjuvant chemotherapy.
Her background was significant for IgA nephropathy, which resulted in progressive chronic kidney disease for 18 years prior to living donor kidney transplant in 2016. Both she and her donor were cytomegalovirus-positive, and she developed cytomegalovirus-associated colitis shortly post-transplantation. Her initial immunosuppression consisted of basiliximab induction followed by tacrolimus, mycophenolic acid, and steroids. She developed antibody-mediated rejection one week post-transplant, which was treated with plasmapheresis, immunoglobulin, and an increase in steroid dose. This was repeated three months later due to biopsy confirming ongoing antibody-mediated rejection. Mycophenolic acid was stopped upon HGSOC diagnosis. She has remained medication-adherent, with regular serum tacrolimus levels within the target range (most recently, 5.3 micrograms/L) and no signs of chronic graft rejection.
Her other comorbidities include diet-controlled, steroid-induced diabetes mellitus; ductal breast carcinoma in situ requiring wide local excision in 2011; asthma; reflux disease; hypertension; and hyperlipidemia. Other medications include prednisone 5 mg daily, acetylsalicylic acid, bisoprolol, trimethoprim–sulfamethoxazole, vitamin D, and inhaled salbutamol as needed.
In December 2019, she developed recurrence in the peritoneum and retroperitoneal lymph nodes, signifying platinum resistance. In January 2020, she commenced weekly paclitaxel 80 mg/m2 with bevacizumab 10 mg/kg every two weeks, and at that time, tacrolimus dose was reduced to 2 mg daily to aim for a serum level of 5 micrograms/L. She continued this therapy for over 6 months, and continued on the same dose of immunosuppression throughout with tacrolimus levels ranging between 3 and 7.6 micrograms/L. Serial imaging and Ca125 confirmed good response to treatment with reduction in size of tumor deposits. Her albumin/creatinine ratio was normal at 0.9 prior to the diagnosis of ovarian cancer, and most recently has been 15.6, signifying microalbuminuria; this has been monitored via urinalysis, which has consistently reported protein as negative or trace. Her most recent estimated glomerular filtration rate was 55 mL/min, similar to pre-diagnosis, and creatinine levels have mostly fluctuated between 95 and 120 umol/L.
Her course has been complicated by grade 2 hypertension (up to 145/95 mmHg) and non-cardiac chest pain, for which amlodipine was switched to ramipril 10 mg daily, and bevacizumab was withheld on two different occasions. Furthermore, she developed a brief period of Kidney Disease: Improving Global Outcomes (KDIGO) stage 1 acute kidney injury (AKI) of pre-renal etiology in September 2020 with creatinine of 160 umol/L, which self-resolved following withdrawal of bevacizumab, and was resumed after a 1-month-long break with no further episodes of kidney injury.
3. Discussion
Nephrotoxicity with angiogenesis inhibitors is relatively common, with proteinuria occurring in over 60% of patients [2]. Most cases are low-grade, transient, and do not require interventions or dose delays; however, more persistent, severe cases presenting as AKI and nephrotic syndrome can occur [2,3]. Risk factors associated with high-grade proteinuria include increased dose, prolonged administration, pre-existing renal disease, and administration of concurrent chemotherapy [4,5].
The pathophysiology of VEGF inhibitor-induced proteinuria remains unclear. Within a normal kidney, VEGF is produced by podocytes, and VEGF receptors are typically present on the glomerular and peritubular endothelium in addition to mesangial cells [6]. Inhibition of VEGF is thought to cause loss of endothelial fenestrations, podocyte injury and reduce endothelial proliferation, ultimately causing disruption of glomerular membranes [6]. Some cases have also demonstrated subacute thrombotic microangiopathy with endotheliosis and membranoproliferative changes [7]. Another manifestation of nephrotoxicity that is commonly seen is hypertension, occurring in more than a third of patients, which arises due to various mechanisms of renal vascular injury including inhibition of nitric oxide, rarefaction of microvasculature, and neuroendocrine dysregulation [8,9]. It is also hypothesized to increase intraglomerular pressure and ultrafiltration, leading to proteinuria [10].
The lack of clarity surrounding pathophysiology of VEGF inhibitor-induced nephrotoxicity [11] is demonstrated by the heterogeneity of published reports on glomerulopathy and other manifestations, including minimal change disease, collapsing glomerulopathy, membranoproliferative glomerulonephritis, focal segmental glomerulosclerosis, cryoglobulinemic glomerulonephritis, acute tubular necrosis, and interstitial nephritis [7]. Furthermore, worsening kidney disease can further exacerbate hypertension, which may perpetuate AKI [10].
Treatment for low-grade proteinuria usually includes an angiotensin-converting enzyme (ACE) inhibitor or an angiotensin receptor blocker to reduce glomerular filtration pressure, and anti-VEGF treatment can be continued providing stable proteinuria. However, proteinuria may worsen to the nephrotic range (>3 g in 24 h) with nephrotic syndrome. Even after cessation of VEGF inhibitor therapy, there are documented cases of persistent proteinuria [12].
3.1. Angiogenesis Inhibitors Post-Transplant
The role of the VEGF pathway in the pathogenesis of post-transplant complications is poorly understood, with preliminary reports across various organ transplants showing hypothesis-generating results [13,14,15,16]. Upregulation of angiogenesis factors was associated with increased allograft vasculopathy, bronchiolitis obliterans, and recurrence of hepatocellular carcinoma in cardiac, pulmonary, and liver transplants, respectively [13,15,16,17]. In renal transplants, VEGF is thought to be upregulated in acute and chronic rejection, particularly associated with cyclosporine [14].
As the first VEGF inhibitor used for anti-cancer treatment, bevacizumab was approved by the United States Food and Drug Administration in 2004, and is now licensed for use in many cancers [18] (Table 1). In ovarian cancer, patients who were prior recipients of solid organ transplants or who were receiving immunosuppressive therapies were not excluded from randomized bevacizumab trials, but those with pre-existing uncontrolled hypertension or renal dysfunction based on serum creatinine ≥ 1.6 mg/dL or proteinuria > 1 g per 24 h were excluded [19,20,21]. Similarly, in other large randomized angiogenesis inhibitor studies across other tumor sites, prior solid organ transplant or use of immunosuppressants is not an exclusion criterion, aside from studies involving immune checkpoint inhibitors [22]. Identification of patients enrolled in large angiogenesis inhibitor trials who had received prior transplants could potentially make for an interesting post-hoc pooled analysis.
Similarly, reports on angiogenesis inhibition in solid organ transplant patients remain scant in the literature, as highlighted by a review on bevacizumab toxicity by Fenoglio et al [9]. Musri et al. reported a case of colorectal cancer post-renal transplantation with baseline proteinuria, which significantly worsened on administration of intravenous 5-fluorouracil, irinotecan, oxaliplatin, and bevacizumab [23]. Cheungpasitporn et al. described two cases with renal allograft dysfunction following administration of intravitreal bevacizumab, aflibercept, or ranibizumab [24]. Doses of anti-angiogenics were lower but not specified within this report. Although neither case proved causality with anti-VEGF therapy, one was diagnosed with phospholipase A2 receptor-negative membranous nephropathy, and the second revealed acute and chronic antibody-mediated rejection with glomerular thrombi and transplant glomerulopathy. Jonkers and Buren reported a case of worsening IgA nephropathy presenting with nephrotic-range proteinuria post-renal transplantation on sorafenib [25]. These reports highlight the potential severe nephrotoxicity known to be associated with angiogenesis inhibitors; however, there remain few documented positive experiences with angiogenesis inhibitor use in the post-transplant setting. Given the prevalence of nephrotoxicity with these agents, these considerations are particularly prudent for renal transplant recipients, but reports in other organ transplants remain similarly scarce.
3.2. Medication Interactions
Solid organ transplant recipients frequently take maintenance immunosuppressive agents, including but not limited to corticosteroids, calcineurin inhibitors, anti-proliferative agents, and mTOR inhibitors, which are associated with various complications and drug interactions (Table 2).
In the case presented, tacrolimus levels were measured every few months to be within the therapeutic range; this is significant as potential drug interactions between transplant medications and angiogenesis inhibitors involve pharmacokinetic and pharmacodynamic interactions. Pharmacokinetic interactions typically occur due to cytochrome P450 enzyme (CYP) and P-glycoprotein (P-gp) drug transport systems [26,27,28], and risk of competitive metabolism as substrates for the same enzyme or transporter may increase serum levels. Axitinib and sorafenib are CYP3A4 and P-gp substrates, and cabozantinib and pazopanib are substrates and inhibitors of both enzyme systems [29,30,31,32]. Drug interactions are well-described for CYP3A4 substrates cyclosporine, tacrolimus, and sirolimus, with metabolic inhibition leading to increased immunosuppressant concentrations (~20%) [28]. Pharmacodynamic interactions primarily concern cumulative toxicities between these two classes (Table 2) [26,27,33,34]. Interestingly, Onodera et al. reported upon a case of metastatic colorectal cancer post-renal transplant where a patient was administered five cycles of 5-fluorouracil, oxaliplatin, and bevacizumab where although severe proteinuria occurred, serum tacrolimus levels were not affected throughout the course of treatment [35]. This report remains one of the only cases in the literature that demonstrates stability of immunosuppression whilst on bevacizumab post-transplant.
3.3. Long-Term Adverse Events
One of the leading causes of morbidity and mortality in solid organ transplant recipients is malignancy, most commonly non-melanomatous skin cancers [36]. Other malignancies such as colorectal, kidney, and cervical cancers are also prevalent in the post-transplant context, and angiogenesis inhibitors such as bevacizumab are commonly used in metastatic disease [18] (Table 1). Surveillance recommendations within transplant recipients are variable across the globe due to a paucity of robust screening trials [37]. Other long-term complications associated with organ transplantation and prolonged immunosuppressant use include cardiovascular disease, diabetes mellitus, hypertension, and infection associated with cytopenia. In the patient presented, adverse events have not outweighed benefits of ongoing treatment, but this will need close monitoring given the risk of overlapping toxicities as long-term adverse event data remain limited [3,38].
4. Conclusions
Whilst there are minimal data justifying that bevacizumab or other angiogenesis inhibitors are unsafe in the post-transplant setting, there is similarly scarce literature demonstrating safe administration, as in the patient presented. As life expectancy continues to improve with increasing indications for transplantation, long-term risks for malignancy with prolonged immunosuppression are increasingly relevant as a cause of mortality in solid organ transplant recipients. In patients who have undergone renal transplantation, careful consideration of treatment options with risk of nephrotoxicity and close monitoring remains paramount.
Although treating oncologists should remain vigilant about potential drug interactions and overlapping toxicities, these are not necessarily contraindications for agents such as bevacizumab. Treatment decisions should consider the best available evidence, and collating information about toxicity and tolerance from randomized trials and post-approval Phase IV studies would provide detailed information from at-risk subgroups. This calls for a stratified, inclusive approach to allow enrolment of those with chronic diseases and comorbidities in prospective trials, allowing objective assessment of the risk–benefit ratio.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Author Contributions
Conceptualization, L.K., A.M.O.; methodology, L.K., J.D.; software, L.K.; validation, L.K., J.D., J.S., A.K.; resources, L.K.; data curation, L.K., S.L., J.S.; writing—original draft preparation, L.K., J.D.; writing—review and editing, L.K., J.D., K.K., A.K., J.S., S.L., A.M.O.; visualization, L.K., K.K.; supervision, S.L., A.M.O. All authors have read and agreed to the published version of the manuscript.
Funding
This research received no external funding.
Institutional Review Board Statement
As this was a report with fewer than three patients, institutional approval was not sought as per the University Health Network Research Ethics Board guidance document on case reports.
Informed Consent Statement
Informed consent was obtained from all subjects involved in the study and written informed consent has been obtained from the patient to publish this paper.
Data Availability Statement
Not applicable.
Conflicts of Interest
We have read and understood Current Oncology’s policy on disclosing conflicts of interest and declare that we have none.
Abbreviations
VEGFR vascular endothelial growth factor receptor
EGFR epidermal growth factor receptor
FGFR fibroblast growth factor receptor
PDGFR platelet-derived growth factor receptor
SCF stem cell factor
PI3K phosphoinositide 3-kinase
AKT protein kinase B
mTOR mammalian target of rapamycin
RET RET proto-oncogene
AXL AXL receptor tyrosine kinase
TAH–BSO total abdominal hysterectomy and bilateral salpingo-oophorectomy
Ca125 cancer antigen 125
U/ml units per milliliter
AKI acute kidney injury
CYP cytochrome P450
HGSOC high-grade serous ovarian carcinoma
FOLFOX4 5-fluorouracil, leucovorin, and oxaliplatin
IFL irinotecan, leucovorin, and 5-fluorouracil
P-gp P-glycoprotein
Figure 1 Angiogenesis pathways in malignancy and targeted therapies commonly used. Monoclonal antibodies including bevacizumab and ramucirumab and tyrosine kinase inhibitors including sunitinib, sorafenib, lenvatinib, pazopanib, axitinib, cediranib, and cabozantinib have been included (this list is not exhaustive). Receptors including VEGFR, c-kit, epidermal growth factor receptor (EGFR), fibroblast growth factor receptor (FGFR), and PDGFR are displayed here, but certain medications also target other pathways not displayed here such as AXL and RET. The VEGFR pathway intersects with multiple cell signaling pathways, including the PI3K/AKT/mTOR pathway. Created with Biorender.com.
Figure 2 (a) Summary oncology treatment timeline. (b) Ca125 trend from diagnosis to current.
curroncol-28-00064-t001_Table 1Table 1 Food and Drug Administration approved indications for bevacizumab [18].
Cancer Stage Usage
Colorectal Metastatic, first-line 5 mg/kg every two weeks with bolus IFL
10 mg/kg every two weeks with FOLFOX4
Metastatic, recurrent after first-line bevacizumab-containing regimen 5 mg/kg every two weeks, or 7.5 mg/kg every three weeks with fluoropyrimidine–irinotecan, or fluoropyrimidine–oxaliplatin-based chemotherapy
Non-squamous, non-small-cell lung Unresectable, locally advanced, recurrent, or metastatic 15 mg/kg every three weeks with carboplatin and paclitaxel
Glioblastoma Recurrent 10 mg/kg every two weeks
Renal cell Metastatic 10 mg/kg every two weeks with interferon-alfa
Cervical Persistent, recurrent, or metastatic 15 mg/kg every three weeks with paclitaxel and cisplatin, or paclitaxel and topotecan
Epithelial ovarian, fallopian tube, or primary peritoneal III or IV, following surgical resection 15 mg/kg every three weeks with carboplatin and paclitaxel for up to six cycles, followed by 15 mg/kg every three weeks as a single agent for up to 22 cycles
Recurrent, platinum-sensitive 15 mg/kg every three weeks with carboplatin and either paclitaxel (6–8 cycles) or gemcitabine (6–10 cycles) followed by 15 mg/kg every 3 weeks as a single agent
Recurrent, platinum-resistant 10 mg/kg every two weeks with paclitaxel, pegylated liposomal doxorubicin, or topotecan given weekly
15 mg/kg every three weeks with topotecan every three weeks
Hepatocellular Unresectable or metastatic, first-line 15 mg/kg with atezolizumab every three weeks
Abbreviations: mg/kg = milligrams per kilogram; IFL = infusional fluoropyrimidine; FOLFOX4 = 5-fluorouracil, folic acid, and oxaliplatin.
curroncol-28-00064-t002_Table 2Table 2 Drug–drug interactions between post-transplant immunosuppressive medications and angiogenesis inhibitors.
Transplant Medication Potential Interactions with Anti-Angiogenesis Agents [26,27,28,29,30,31,32,33,34]
Cyclosporine
Increased cyclosporine levels and subsequent toxicity due to CYP3A4 and P-gp-mediated drug interactions (e.g., cabozatinib, axitinib, pazopanib, sorafenib, sunitinib)
Tacrolimus
Increased tacrolimus levels and subsequent toxicity due to inhibition of or competition with CYP3A4 metabolism and P-gp-mediated transport (e.g., cabozatinib, axitinib, pazopanib, sorafenib, sunitinib)
Additive impairment of the renal function (e.g., cediranib, axitinib, pazopanib)
Increased risk of QT prolongation with other agents that prolong the QT interval (e.g., cabozantinib, pazopanib, sorafenib, sunitinib)
Mycophenolate mofetil
Exaggerated leukopenia (e.g., ramucirumab, bevacizumab, sunitinib)
Azathioprine
Exaggerated leukopenia (e.g., ramucirumab, bevacizumab, sunitinib)
Sirolimus
Increased tacrolimus levels and subsequent toxicity due to inhibition of or competition with CYP3A4 metabolism and P-gp-mediated transport (e.g., cabozatinib, axitinib, pazopanib, sorafenib, sunitinib)
Additive impairment of the renal function (e.g., cediranib, axitinib, pazopanib)
Additive impairment of wound healing
Everolimus
Increased everolimus levels and subsequent toxicity due to inhibition of CYP3A4 metabolism and P-gp-mediated transport (e.g., cabozantinib, pazopanib)
Additive impairment of the renal function (e.g., cediranib, axitinib, pazopanib)
Additive impairment of wound healing
Corticosteroids
Competitive CYP3A4 metabolism (e.g., prednisone) with other CYP3A4 substrates (e.g., cabozantinib, axitinib, pazopanib, sorafenib, sunitinib)
Abbreviations: CYP3A4 = cytochrome P450 3A4; P-gp = P-glycoprotein. | ALBUTEROL, ASPIRIN, BEVACIZUMAB, BISOPROLOL, PACLITAXEL, PREDNISONE, SULFAMETHOXAZOLE\TRIMETHOPRIM, TACROLIMUS, VITAMIN D NOS | DrugsGivenReaction | CC BY | 33499164 | 18,991,561 | 2021-01-22 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Ovarian cancer'. | Angiogenesis Inhibitors as Anti-Cancer Therapy Following Renal Transplantation: A Case Report and Review of the Literature.
Solid organ transplant recipients on long-term immunosuppressive medication are at increased risk of developing malignancy, and treatment of advanced cancers with angiogenesis inhibitors in this context has not been widely studied. We present a case of recurrent high-grade serous ovarian carcinoma treated with paclitaxel and bevacizumab in the context of prior renal transplantation where the patient responded well to treatment with controlled toxicities, discussing the potential for increased rates of adverse events and drug interactions in this select population.
1. Introduction
Angiogenesis inhibitors such as vascular endothelial growth factor (VEGF) monoclonal antibodies and tyrosine kinase inhibitors (Figure 1) are standard treatments across various cancer subtypes. In advanced high-grade serous ovarian cancer (HGSOC), bevacizumab is part of the standard of care as maintenance therapy in front-line and recurrent disease [1]. Although the toxicity profile is generally well-managed, nephrotoxicity manifesting as proteinuria remains an important adverse event that requires close monitoring [2]. There is limited literature surrounding angiogenesis inhibitors as anti-cancer treatment in patients who have received solid organ transplants, and thus this report presents a case demonstrating safety of angiogenesis inhibition as anti-cancer therapy in a patient with a stable renal transplant.
2. Case Description
Consent: Fully informed, voluntary, written consent has been obtained to include patient information and publish this report.
Ethics: As this is a case report with fewer than three patients, institutional approval was not sought as per the University Health Network Research Ethics Board guidance document on case reports.
A 47-year-old woman presented with several months of dyspnea and abdominal distension to Princess Margaret Cancer Centre in February 2019 (Figure 2). CT revealed a 13.9 cm pelvic mass with peritoneal carcinomatosis and ascites. Omental biopsy confirmed HGSOC, and she received neoadjuvant platinum-based chemotherapy with excellent tolerance and no renal complications before proceeding to interval debulking in June 2019. There was no visible residual disease, and diagnosis of HGSOC was confirmed, germline and somatic BRCA wild type. She completed three cycles of adjuvant chemotherapy.
Her background was significant for IgA nephropathy, which resulted in progressive chronic kidney disease for 18 years prior to living donor kidney transplant in 2016. Both she and her donor were cytomegalovirus-positive, and she developed cytomegalovirus-associated colitis shortly post-transplantation. Her initial immunosuppression consisted of basiliximab induction followed by tacrolimus, mycophenolic acid, and steroids. She developed antibody-mediated rejection one week post-transplant, which was treated with plasmapheresis, immunoglobulin, and an increase in steroid dose. This was repeated three months later due to biopsy confirming ongoing antibody-mediated rejection. Mycophenolic acid was stopped upon HGSOC diagnosis. She has remained medication-adherent, with regular serum tacrolimus levels within the target range (most recently, 5.3 micrograms/L) and no signs of chronic graft rejection.
Her other comorbidities include diet-controlled, steroid-induced diabetes mellitus; ductal breast carcinoma in situ requiring wide local excision in 2011; asthma; reflux disease; hypertension; and hyperlipidemia. Other medications include prednisone 5 mg daily, acetylsalicylic acid, bisoprolol, trimethoprim–sulfamethoxazole, vitamin D, and inhaled salbutamol as needed.
In December 2019, she developed recurrence in the peritoneum and retroperitoneal lymph nodes, signifying platinum resistance. In January 2020, she commenced weekly paclitaxel 80 mg/m2 with bevacizumab 10 mg/kg every two weeks, and at that time, tacrolimus dose was reduced to 2 mg daily to aim for a serum level of 5 micrograms/L. She continued this therapy for over 6 months, and continued on the same dose of immunosuppression throughout with tacrolimus levels ranging between 3 and 7.6 micrograms/L. Serial imaging and Ca125 confirmed good response to treatment with reduction in size of tumor deposits. Her albumin/creatinine ratio was normal at 0.9 prior to the diagnosis of ovarian cancer, and most recently has been 15.6, signifying microalbuminuria; this has been monitored via urinalysis, which has consistently reported protein as negative or trace. Her most recent estimated glomerular filtration rate was 55 mL/min, similar to pre-diagnosis, and creatinine levels have mostly fluctuated between 95 and 120 umol/L.
Her course has been complicated by grade 2 hypertension (up to 145/95 mmHg) and non-cardiac chest pain, for which amlodipine was switched to ramipril 10 mg daily, and bevacizumab was withheld on two different occasions. Furthermore, she developed a brief period of Kidney Disease: Improving Global Outcomes (KDIGO) stage 1 acute kidney injury (AKI) of pre-renal etiology in September 2020 with creatinine of 160 umol/L, which self-resolved following withdrawal of bevacizumab, and was resumed after a 1-month-long break with no further episodes of kidney injury.
3. Discussion
Nephrotoxicity with angiogenesis inhibitors is relatively common, with proteinuria occurring in over 60% of patients [2]. Most cases are low-grade, transient, and do not require interventions or dose delays; however, more persistent, severe cases presenting as AKI and nephrotic syndrome can occur [2,3]. Risk factors associated with high-grade proteinuria include increased dose, prolonged administration, pre-existing renal disease, and administration of concurrent chemotherapy [4,5].
The pathophysiology of VEGF inhibitor-induced proteinuria remains unclear. Within a normal kidney, VEGF is produced by podocytes, and VEGF receptors are typically present on the glomerular and peritubular endothelium in addition to mesangial cells [6]. Inhibition of VEGF is thought to cause loss of endothelial fenestrations, podocyte injury and reduce endothelial proliferation, ultimately causing disruption of glomerular membranes [6]. Some cases have also demonstrated subacute thrombotic microangiopathy with endotheliosis and membranoproliferative changes [7]. Another manifestation of nephrotoxicity that is commonly seen is hypertension, occurring in more than a third of patients, which arises due to various mechanisms of renal vascular injury including inhibition of nitric oxide, rarefaction of microvasculature, and neuroendocrine dysregulation [8,9]. It is also hypothesized to increase intraglomerular pressure and ultrafiltration, leading to proteinuria [10].
The lack of clarity surrounding pathophysiology of VEGF inhibitor-induced nephrotoxicity [11] is demonstrated by the heterogeneity of published reports on glomerulopathy and other manifestations, including minimal change disease, collapsing glomerulopathy, membranoproliferative glomerulonephritis, focal segmental glomerulosclerosis, cryoglobulinemic glomerulonephritis, acute tubular necrosis, and interstitial nephritis [7]. Furthermore, worsening kidney disease can further exacerbate hypertension, which may perpetuate AKI [10].
Treatment for low-grade proteinuria usually includes an angiotensin-converting enzyme (ACE) inhibitor or an angiotensin receptor blocker to reduce glomerular filtration pressure, and anti-VEGF treatment can be continued providing stable proteinuria. However, proteinuria may worsen to the nephrotic range (>3 g in 24 h) with nephrotic syndrome. Even after cessation of VEGF inhibitor therapy, there are documented cases of persistent proteinuria [12].
3.1. Angiogenesis Inhibitors Post-Transplant
The role of the VEGF pathway in the pathogenesis of post-transplant complications is poorly understood, with preliminary reports across various organ transplants showing hypothesis-generating results [13,14,15,16]. Upregulation of angiogenesis factors was associated with increased allograft vasculopathy, bronchiolitis obliterans, and recurrence of hepatocellular carcinoma in cardiac, pulmonary, and liver transplants, respectively [13,15,16,17]. In renal transplants, VEGF is thought to be upregulated in acute and chronic rejection, particularly associated with cyclosporine [14].
As the first VEGF inhibitor used for anti-cancer treatment, bevacizumab was approved by the United States Food and Drug Administration in 2004, and is now licensed for use in many cancers [18] (Table 1). In ovarian cancer, patients who were prior recipients of solid organ transplants or who were receiving immunosuppressive therapies were not excluded from randomized bevacizumab trials, but those with pre-existing uncontrolled hypertension or renal dysfunction based on serum creatinine ≥ 1.6 mg/dL or proteinuria > 1 g per 24 h were excluded [19,20,21]. Similarly, in other large randomized angiogenesis inhibitor studies across other tumor sites, prior solid organ transplant or use of immunosuppressants is not an exclusion criterion, aside from studies involving immune checkpoint inhibitors [22]. Identification of patients enrolled in large angiogenesis inhibitor trials who had received prior transplants could potentially make for an interesting post-hoc pooled analysis.
Similarly, reports on angiogenesis inhibition in solid organ transplant patients remain scant in the literature, as highlighted by a review on bevacizumab toxicity by Fenoglio et al [9]. Musri et al. reported a case of colorectal cancer post-renal transplantation with baseline proteinuria, which significantly worsened on administration of intravenous 5-fluorouracil, irinotecan, oxaliplatin, and bevacizumab [23]. Cheungpasitporn et al. described two cases with renal allograft dysfunction following administration of intravitreal bevacizumab, aflibercept, or ranibizumab [24]. Doses of anti-angiogenics were lower but not specified within this report. Although neither case proved causality with anti-VEGF therapy, one was diagnosed with phospholipase A2 receptor-negative membranous nephropathy, and the second revealed acute and chronic antibody-mediated rejection with glomerular thrombi and transplant glomerulopathy. Jonkers and Buren reported a case of worsening IgA nephropathy presenting with nephrotic-range proteinuria post-renal transplantation on sorafenib [25]. These reports highlight the potential severe nephrotoxicity known to be associated with angiogenesis inhibitors; however, there remain few documented positive experiences with angiogenesis inhibitor use in the post-transplant setting. Given the prevalence of nephrotoxicity with these agents, these considerations are particularly prudent for renal transplant recipients, but reports in other organ transplants remain similarly scarce.
3.2. Medication Interactions
Solid organ transplant recipients frequently take maintenance immunosuppressive agents, including but not limited to corticosteroids, calcineurin inhibitors, anti-proliferative agents, and mTOR inhibitors, which are associated with various complications and drug interactions (Table 2).
In the case presented, tacrolimus levels were measured every few months to be within the therapeutic range; this is significant as potential drug interactions between transplant medications and angiogenesis inhibitors involve pharmacokinetic and pharmacodynamic interactions. Pharmacokinetic interactions typically occur due to cytochrome P450 enzyme (CYP) and P-glycoprotein (P-gp) drug transport systems [26,27,28], and risk of competitive metabolism as substrates for the same enzyme or transporter may increase serum levels. Axitinib and sorafenib are CYP3A4 and P-gp substrates, and cabozantinib and pazopanib are substrates and inhibitors of both enzyme systems [29,30,31,32]. Drug interactions are well-described for CYP3A4 substrates cyclosporine, tacrolimus, and sirolimus, with metabolic inhibition leading to increased immunosuppressant concentrations (~20%) [28]. Pharmacodynamic interactions primarily concern cumulative toxicities between these two classes (Table 2) [26,27,33,34]. Interestingly, Onodera et al. reported upon a case of metastatic colorectal cancer post-renal transplant where a patient was administered five cycles of 5-fluorouracil, oxaliplatin, and bevacizumab where although severe proteinuria occurred, serum tacrolimus levels were not affected throughout the course of treatment [35]. This report remains one of the only cases in the literature that demonstrates stability of immunosuppression whilst on bevacizumab post-transplant.
3.3. Long-Term Adverse Events
One of the leading causes of morbidity and mortality in solid organ transplant recipients is malignancy, most commonly non-melanomatous skin cancers [36]. Other malignancies such as colorectal, kidney, and cervical cancers are also prevalent in the post-transplant context, and angiogenesis inhibitors such as bevacizumab are commonly used in metastatic disease [18] (Table 1). Surveillance recommendations within transplant recipients are variable across the globe due to a paucity of robust screening trials [37]. Other long-term complications associated with organ transplantation and prolonged immunosuppressant use include cardiovascular disease, diabetes mellitus, hypertension, and infection associated with cytopenia. In the patient presented, adverse events have not outweighed benefits of ongoing treatment, but this will need close monitoring given the risk of overlapping toxicities as long-term adverse event data remain limited [3,38].
4. Conclusions
Whilst there are minimal data justifying that bevacizumab or other angiogenesis inhibitors are unsafe in the post-transplant setting, there is similarly scarce literature demonstrating safe administration, as in the patient presented. As life expectancy continues to improve with increasing indications for transplantation, long-term risks for malignancy with prolonged immunosuppression are increasingly relevant as a cause of mortality in solid organ transplant recipients. In patients who have undergone renal transplantation, careful consideration of treatment options with risk of nephrotoxicity and close monitoring remains paramount.
Although treating oncologists should remain vigilant about potential drug interactions and overlapping toxicities, these are not necessarily contraindications for agents such as bevacizumab. Treatment decisions should consider the best available evidence, and collating information about toxicity and tolerance from randomized trials and post-approval Phase IV studies would provide detailed information from at-risk subgroups. This calls for a stratified, inclusive approach to allow enrolment of those with chronic diseases and comorbidities in prospective trials, allowing objective assessment of the risk–benefit ratio.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Author Contributions
Conceptualization, L.K., A.M.O.; methodology, L.K., J.D.; software, L.K.; validation, L.K., J.D., J.S., A.K.; resources, L.K.; data curation, L.K., S.L., J.S.; writing—original draft preparation, L.K., J.D.; writing—review and editing, L.K., J.D., K.K., A.K., J.S., S.L., A.M.O.; visualization, L.K., K.K.; supervision, S.L., A.M.O. All authors have read and agreed to the published version of the manuscript.
Funding
This research received no external funding.
Institutional Review Board Statement
As this was a report with fewer than three patients, institutional approval was not sought as per the University Health Network Research Ethics Board guidance document on case reports.
Informed Consent Statement
Informed consent was obtained from all subjects involved in the study and written informed consent has been obtained from the patient to publish this paper.
Data Availability Statement
Not applicable.
Conflicts of Interest
We have read and understood Current Oncology’s policy on disclosing conflicts of interest and declare that we have none.
Abbreviations
VEGFR vascular endothelial growth factor receptor
EGFR epidermal growth factor receptor
FGFR fibroblast growth factor receptor
PDGFR platelet-derived growth factor receptor
SCF stem cell factor
PI3K phosphoinositide 3-kinase
AKT protein kinase B
mTOR mammalian target of rapamycin
RET RET proto-oncogene
AXL AXL receptor tyrosine kinase
TAH–BSO total abdominal hysterectomy and bilateral salpingo-oophorectomy
Ca125 cancer antigen 125
U/ml units per milliliter
AKI acute kidney injury
CYP cytochrome P450
HGSOC high-grade serous ovarian carcinoma
FOLFOX4 5-fluorouracil, leucovorin, and oxaliplatin
IFL irinotecan, leucovorin, and 5-fluorouracil
P-gp P-glycoprotein
Figure 1 Angiogenesis pathways in malignancy and targeted therapies commonly used. Monoclonal antibodies including bevacizumab and ramucirumab and tyrosine kinase inhibitors including sunitinib, sorafenib, lenvatinib, pazopanib, axitinib, cediranib, and cabozantinib have been included (this list is not exhaustive). Receptors including VEGFR, c-kit, epidermal growth factor receptor (EGFR), fibroblast growth factor receptor (FGFR), and PDGFR are displayed here, but certain medications also target other pathways not displayed here such as AXL and RET. The VEGFR pathway intersects with multiple cell signaling pathways, including the PI3K/AKT/mTOR pathway. Created with Biorender.com.
Figure 2 (a) Summary oncology treatment timeline. (b) Ca125 trend from diagnosis to current.
curroncol-28-00064-t001_Table 1Table 1 Food and Drug Administration approved indications for bevacizumab [18].
Cancer Stage Usage
Colorectal Metastatic, first-line 5 mg/kg every two weeks with bolus IFL
10 mg/kg every two weeks with FOLFOX4
Metastatic, recurrent after first-line bevacizumab-containing regimen 5 mg/kg every two weeks, or 7.5 mg/kg every three weeks with fluoropyrimidine–irinotecan, or fluoropyrimidine–oxaliplatin-based chemotherapy
Non-squamous, non-small-cell lung Unresectable, locally advanced, recurrent, or metastatic 15 mg/kg every three weeks with carboplatin and paclitaxel
Glioblastoma Recurrent 10 mg/kg every two weeks
Renal cell Metastatic 10 mg/kg every two weeks with interferon-alfa
Cervical Persistent, recurrent, or metastatic 15 mg/kg every three weeks with paclitaxel and cisplatin, or paclitaxel and topotecan
Epithelial ovarian, fallopian tube, or primary peritoneal III or IV, following surgical resection 15 mg/kg every three weeks with carboplatin and paclitaxel for up to six cycles, followed by 15 mg/kg every three weeks as a single agent for up to 22 cycles
Recurrent, platinum-sensitive 15 mg/kg every three weeks with carboplatin and either paclitaxel (6–8 cycles) or gemcitabine (6–10 cycles) followed by 15 mg/kg every 3 weeks as a single agent
Recurrent, platinum-resistant 10 mg/kg every two weeks with paclitaxel, pegylated liposomal doxorubicin, or topotecan given weekly
15 mg/kg every three weeks with topotecan every three weeks
Hepatocellular Unresectable or metastatic, first-line 15 mg/kg with atezolizumab every three weeks
Abbreviations: mg/kg = milligrams per kilogram; IFL = infusional fluoropyrimidine; FOLFOX4 = 5-fluorouracil, folic acid, and oxaliplatin.
curroncol-28-00064-t002_Table 2Table 2 Drug–drug interactions between post-transplant immunosuppressive medications and angiogenesis inhibitors.
Transplant Medication Potential Interactions with Anti-Angiogenesis Agents [26,27,28,29,30,31,32,33,34]
Cyclosporine
Increased cyclosporine levels and subsequent toxicity due to CYP3A4 and P-gp-mediated drug interactions (e.g., cabozatinib, axitinib, pazopanib, sorafenib, sunitinib)
Tacrolimus
Increased tacrolimus levels and subsequent toxicity due to inhibition of or competition with CYP3A4 metabolism and P-gp-mediated transport (e.g., cabozatinib, axitinib, pazopanib, sorafenib, sunitinib)
Additive impairment of the renal function (e.g., cediranib, axitinib, pazopanib)
Increased risk of QT prolongation with other agents that prolong the QT interval (e.g., cabozantinib, pazopanib, sorafenib, sunitinib)
Mycophenolate mofetil
Exaggerated leukopenia (e.g., ramucirumab, bevacizumab, sunitinib)
Azathioprine
Exaggerated leukopenia (e.g., ramucirumab, bevacizumab, sunitinib)
Sirolimus
Increased tacrolimus levels and subsequent toxicity due to inhibition of or competition with CYP3A4 metabolism and P-gp-mediated transport (e.g., cabozatinib, axitinib, pazopanib, sorafenib, sunitinib)
Additive impairment of the renal function (e.g., cediranib, axitinib, pazopanib)
Additive impairment of wound healing
Everolimus
Increased everolimus levels and subsequent toxicity due to inhibition of CYP3A4 metabolism and P-gp-mediated transport (e.g., cabozantinib, pazopanib)
Additive impairment of the renal function (e.g., cediranib, axitinib, pazopanib)
Additive impairment of wound healing
Corticosteroids
Competitive CYP3A4 metabolism (e.g., prednisone) with other CYP3A4 substrates (e.g., cabozantinib, axitinib, pazopanib, sorafenib, sunitinib)
Abbreviations: CYP3A4 = cytochrome P450 3A4; P-gp = P-glycoprotein. | ALBUTEROL, ASPIRIN, BASILIXIMAB, BEVACIZUMAB, BISOPROLOL, MYCOPHENOLIC ACID, PACLITAXEL, PREDNISONE, SULFAMETHOXAZOLE\TRIMETHOPRIM, TACROLIMUS, VITAMIN D NOS | DrugsGivenReaction | CC BY | 33499164 | 19,169,975 | 2021-01-22 |
What was the administration route of drug 'ALBUTEROL'? | Angiogenesis Inhibitors as Anti-Cancer Therapy Following Renal Transplantation: A Case Report and Review of the Literature.
Solid organ transplant recipients on long-term immunosuppressive medication are at increased risk of developing malignancy, and treatment of advanced cancers with angiogenesis inhibitors in this context has not been widely studied. We present a case of recurrent high-grade serous ovarian carcinoma treated with paclitaxel and bevacizumab in the context of prior renal transplantation where the patient responded well to treatment with controlled toxicities, discussing the potential for increased rates of adverse events and drug interactions in this select population.
1. Introduction
Angiogenesis inhibitors such as vascular endothelial growth factor (VEGF) monoclonal antibodies and tyrosine kinase inhibitors (Figure 1) are standard treatments across various cancer subtypes. In advanced high-grade serous ovarian cancer (HGSOC), bevacizumab is part of the standard of care as maintenance therapy in front-line and recurrent disease [1]. Although the toxicity profile is generally well-managed, nephrotoxicity manifesting as proteinuria remains an important adverse event that requires close monitoring [2]. There is limited literature surrounding angiogenesis inhibitors as anti-cancer treatment in patients who have received solid organ transplants, and thus this report presents a case demonstrating safety of angiogenesis inhibition as anti-cancer therapy in a patient with a stable renal transplant.
2. Case Description
Consent: Fully informed, voluntary, written consent has been obtained to include patient information and publish this report.
Ethics: As this is a case report with fewer than three patients, institutional approval was not sought as per the University Health Network Research Ethics Board guidance document on case reports.
A 47-year-old woman presented with several months of dyspnea and abdominal distension to Princess Margaret Cancer Centre in February 2019 (Figure 2). CT revealed a 13.9 cm pelvic mass with peritoneal carcinomatosis and ascites. Omental biopsy confirmed HGSOC, and she received neoadjuvant platinum-based chemotherapy with excellent tolerance and no renal complications before proceeding to interval debulking in June 2019. There was no visible residual disease, and diagnosis of HGSOC was confirmed, germline and somatic BRCA wild type. She completed three cycles of adjuvant chemotherapy.
Her background was significant for IgA nephropathy, which resulted in progressive chronic kidney disease for 18 years prior to living donor kidney transplant in 2016. Both she and her donor were cytomegalovirus-positive, and she developed cytomegalovirus-associated colitis shortly post-transplantation. Her initial immunosuppression consisted of basiliximab induction followed by tacrolimus, mycophenolic acid, and steroids. She developed antibody-mediated rejection one week post-transplant, which was treated with plasmapheresis, immunoglobulin, and an increase in steroid dose. This was repeated three months later due to biopsy confirming ongoing antibody-mediated rejection. Mycophenolic acid was stopped upon HGSOC diagnosis. She has remained medication-adherent, with regular serum tacrolimus levels within the target range (most recently, 5.3 micrograms/L) and no signs of chronic graft rejection.
Her other comorbidities include diet-controlled, steroid-induced diabetes mellitus; ductal breast carcinoma in situ requiring wide local excision in 2011; asthma; reflux disease; hypertension; and hyperlipidemia. Other medications include prednisone 5 mg daily, acetylsalicylic acid, bisoprolol, trimethoprim–sulfamethoxazole, vitamin D, and inhaled salbutamol as needed.
In December 2019, she developed recurrence in the peritoneum and retroperitoneal lymph nodes, signifying platinum resistance. In January 2020, she commenced weekly paclitaxel 80 mg/m2 with bevacizumab 10 mg/kg every two weeks, and at that time, tacrolimus dose was reduced to 2 mg daily to aim for a serum level of 5 micrograms/L. She continued this therapy for over 6 months, and continued on the same dose of immunosuppression throughout with tacrolimus levels ranging between 3 and 7.6 micrograms/L. Serial imaging and Ca125 confirmed good response to treatment with reduction in size of tumor deposits. Her albumin/creatinine ratio was normal at 0.9 prior to the diagnosis of ovarian cancer, and most recently has been 15.6, signifying microalbuminuria; this has been monitored via urinalysis, which has consistently reported protein as negative or trace. Her most recent estimated glomerular filtration rate was 55 mL/min, similar to pre-diagnosis, and creatinine levels have mostly fluctuated between 95 and 120 umol/L.
Her course has been complicated by grade 2 hypertension (up to 145/95 mmHg) and non-cardiac chest pain, for which amlodipine was switched to ramipril 10 mg daily, and bevacizumab was withheld on two different occasions. Furthermore, she developed a brief period of Kidney Disease: Improving Global Outcomes (KDIGO) stage 1 acute kidney injury (AKI) of pre-renal etiology in September 2020 with creatinine of 160 umol/L, which self-resolved following withdrawal of bevacizumab, and was resumed after a 1-month-long break with no further episodes of kidney injury.
3. Discussion
Nephrotoxicity with angiogenesis inhibitors is relatively common, with proteinuria occurring in over 60% of patients [2]. Most cases are low-grade, transient, and do not require interventions or dose delays; however, more persistent, severe cases presenting as AKI and nephrotic syndrome can occur [2,3]. Risk factors associated with high-grade proteinuria include increased dose, prolonged administration, pre-existing renal disease, and administration of concurrent chemotherapy [4,5].
The pathophysiology of VEGF inhibitor-induced proteinuria remains unclear. Within a normal kidney, VEGF is produced by podocytes, and VEGF receptors are typically present on the glomerular and peritubular endothelium in addition to mesangial cells [6]. Inhibition of VEGF is thought to cause loss of endothelial fenestrations, podocyte injury and reduce endothelial proliferation, ultimately causing disruption of glomerular membranes [6]. Some cases have also demonstrated subacute thrombotic microangiopathy with endotheliosis and membranoproliferative changes [7]. Another manifestation of nephrotoxicity that is commonly seen is hypertension, occurring in more than a third of patients, which arises due to various mechanisms of renal vascular injury including inhibition of nitric oxide, rarefaction of microvasculature, and neuroendocrine dysregulation [8,9]. It is also hypothesized to increase intraglomerular pressure and ultrafiltration, leading to proteinuria [10].
The lack of clarity surrounding pathophysiology of VEGF inhibitor-induced nephrotoxicity [11] is demonstrated by the heterogeneity of published reports on glomerulopathy and other manifestations, including minimal change disease, collapsing glomerulopathy, membranoproliferative glomerulonephritis, focal segmental glomerulosclerosis, cryoglobulinemic glomerulonephritis, acute tubular necrosis, and interstitial nephritis [7]. Furthermore, worsening kidney disease can further exacerbate hypertension, which may perpetuate AKI [10].
Treatment for low-grade proteinuria usually includes an angiotensin-converting enzyme (ACE) inhibitor or an angiotensin receptor blocker to reduce glomerular filtration pressure, and anti-VEGF treatment can be continued providing stable proteinuria. However, proteinuria may worsen to the nephrotic range (>3 g in 24 h) with nephrotic syndrome. Even after cessation of VEGF inhibitor therapy, there are documented cases of persistent proteinuria [12].
3.1. Angiogenesis Inhibitors Post-Transplant
The role of the VEGF pathway in the pathogenesis of post-transplant complications is poorly understood, with preliminary reports across various organ transplants showing hypothesis-generating results [13,14,15,16]. Upregulation of angiogenesis factors was associated with increased allograft vasculopathy, bronchiolitis obliterans, and recurrence of hepatocellular carcinoma in cardiac, pulmonary, and liver transplants, respectively [13,15,16,17]. In renal transplants, VEGF is thought to be upregulated in acute and chronic rejection, particularly associated with cyclosporine [14].
As the first VEGF inhibitor used for anti-cancer treatment, bevacizumab was approved by the United States Food and Drug Administration in 2004, and is now licensed for use in many cancers [18] (Table 1). In ovarian cancer, patients who were prior recipients of solid organ transplants or who were receiving immunosuppressive therapies were not excluded from randomized bevacizumab trials, but those with pre-existing uncontrolled hypertension or renal dysfunction based on serum creatinine ≥ 1.6 mg/dL or proteinuria > 1 g per 24 h were excluded [19,20,21]. Similarly, in other large randomized angiogenesis inhibitor studies across other tumor sites, prior solid organ transplant or use of immunosuppressants is not an exclusion criterion, aside from studies involving immune checkpoint inhibitors [22]. Identification of patients enrolled in large angiogenesis inhibitor trials who had received prior transplants could potentially make for an interesting post-hoc pooled analysis.
Similarly, reports on angiogenesis inhibition in solid organ transplant patients remain scant in the literature, as highlighted by a review on bevacizumab toxicity by Fenoglio et al [9]. Musri et al. reported a case of colorectal cancer post-renal transplantation with baseline proteinuria, which significantly worsened on administration of intravenous 5-fluorouracil, irinotecan, oxaliplatin, and bevacizumab [23]. Cheungpasitporn et al. described two cases with renal allograft dysfunction following administration of intravitreal bevacizumab, aflibercept, or ranibizumab [24]. Doses of anti-angiogenics were lower but not specified within this report. Although neither case proved causality with anti-VEGF therapy, one was diagnosed with phospholipase A2 receptor-negative membranous nephropathy, and the second revealed acute and chronic antibody-mediated rejection with glomerular thrombi and transplant glomerulopathy. Jonkers and Buren reported a case of worsening IgA nephropathy presenting with nephrotic-range proteinuria post-renal transplantation on sorafenib [25]. These reports highlight the potential severe nephrotoxicity known to be associated with angiogenesis inhibitors; however, there remain few documented positive experiences with angiogenesis inhibitor use in the post-transplant setting. Given the prevalence of nephrotoxicity with these agents, these considerations are particularly prudent for renal transplant recipients, but reports in other organ transplants remain similarly scarce.
3.2. Medication Interactions
Solid organ transplant recipients frequently take maintenance immunosuppressive agents, including but not limited to corticosteroids, calcineurin inhibitors, anti-proliferative agents, and mTOR inhibitors, which are associated with various complications and drug interactions (Table 2).
In the case presented, tacrolimus levels were measured every few months to be within the therapeutic range; this is significant as potential drug interactions between transplant medications and angiogenesis inhibitors involve pharmacokinetic and pharmacodynamic interactions. Pharmacokinetic interactions typically occur due to cytochrome P450 enzyme (CYP) and P-glycoprotein (P-gp) drug transport systems [26,27,28], and risk of competitive metabolism as substrates for the same enzyme or transporter may increase serum levels. Axitinib and sorafenib are CYP3A4 and P-gp substrates, and cabozantinib and pazopanib are substrates and inhibitors of both enzyme systems [29,30,31,32]. Drug interactions are well-described for CYP3A4 substrates cyclosporine, tacrolimus, and sirolimus, with metabolic inhibition leading to increased immunosuppressant concentrations (~20%) [28]. Pharmacodynamic interactions primarily concern cumulative toxicities between these two classes (Table 2) [26,27,33,34]. Interestingly, Onodera et al. reported upon a case of metastatic colorectal cancer post-renal transplant where a patient was administered five cycles of 5-fluorouracil, oxaliplatin, and bevacizumab where although severe proteinuria occurred, serum tacrolimus levels were not affected throughout the course of treatment [35]. This report remains one of the only cases in the literature that demonstrates stability of immunosuppression whilst on bevacizumab post-transplant.
3.3. Long-Term Adverse Events
One of the leading causes of morbidity and mortality in solid organ transplant recipients is malignancy, most commonly non-melanomatous skin cancers [36]. Other malignancies such as colorectal, kidney, and cervical cancers are also prevalent in the post-transplant context, and angiogenesis inhibitors such as bevacizumab are commonly used in metastatic disease [18] (Table 1). Surveillance recommendations within transplant recipients are variable across the globe due to a paucity of robust screening trials [37]. Other long-term complications associated with organ transplantation and prolonged immunosuppressant use include cardiovascular disease, diabetes mellitus, hypertension, and infection associated with cytopenia. In the patient presented, adverse events have not outweighed benefits of ongoing treatment, but this will need close monitoring given the risk of overlapping toxicities as long-term adverse event data remain limited [3,38].
4. Conclusions
Whilst there are minimal data justifying that bevacizumab or other angiogenesis inhibitors are unsafe in the post-transplant setting, there is similarly scarce literature demonstrating safe administration, as in the patient presented. As life expectancy continues to improve with increasing indications for transplantation, long-term risks for malignancy with prolonged immunosuppression are increasingly relevant as a cause of mortality in solid organ transplant recipients. In patients who have undergone renal transplantation, careful consideration of treatment options with risk of nephrotoxicity and close monitoring remains paramount.
Although treating oncologists should remain vigilant about potential drug interactions and overlapping toxicities, these are not necessarily contraindications for agents such as bevacizumab. Treatment decisions should consider the best available evidence, and collating information about toxicity and tolerance from randomized trials and post-approval Phase IV studies would provide detailed information from at-risk subgroups. This calls for a stratified, inclusive approach to allow enrolment of those with chronic diseases and comorbidities in prospective trials, allowing objective assessment of the risk–benefit ratio.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Author Contributions
Conceptualization, L.K., A.M.O.; methodology, L.K., J.D.; software, L.K.; validation, L.K., J.D., J.S., A.K.; resources, L.K.; data curation, L.K., S.L., J.S.; writing—original draft preparation, L.K., J.D.; writing—review and editing, L.K., J.D., K.K., A.K., J.S., S.L., A.M.O.; visualization, L.K., K.K.; supervision, S.L., A.M.O. All authors have read and agreed to the published version of the manuscript.
Funding
This research received no external funding.
Institutional Review Board Statement
As this was a report with fewer than three patients, institutional approval was not sought as per the University Health Network Research Ethics Board guidance document on case reports.
Informed Consent Statement
Informed consent was obtained from all subjects involved in the study and written informed consent has been obtained from the patient to publish this paper.
Data Availability Statement
Not applicable.
Conflicts of Interest
We have read and understood Current Oncology’s policy on disclosing conflicts of interest and declare that we have none.
Abbreviations
VEGFR vascular endothelial growth factor receptor
EGFR epidermal growth factor receptor
FGFR fibroblast growth factor receptor
PDGFR platelet-derived growth factor receptor
SCF stem cell factor
PI3K phosphoinositide 3-kinase
AKT protein kinase B
mTOR mammalian target of rapamycin
RET RET proto-oncogene
AXL AXL receptor tyrosine kinase
TAH–BSO total abdominal hysterectomy and bilateral salpingo-oophorectomy
Ca125 cancer antigen 125
U/ml units per milliliter
AKI acute kidney injury
CYP cytochrome P450
HGSOC high-grade serous ovarian carcinoma
FOLFOX4 5-fluorouracil, leucovorin, and oxaliplatin
IFL irinotecan, leucovorin, and 5-fluorouracil
P-gp P-glycoprotein
Figure 1 Angiogenesis pathways in malignancy and targeted therapies commonly used. Monoclonal antibodies including bevacizumab and ramucirumab and tyrosine kinase inhibitors including sunitinib, sorafenib, lenvatinib, pazopanib, axitinib, cediranib, and cabozantinib have been included (this list is not exhaustive). Receptors including VEGFR, c-kit, epidermal growth factor receptor (EGFR), fibroblast growth factor receptor (FGFR), and PDGFR are displayed here, but certain medications also target other pathways not displayed here such as AXL and RET. The VEGFR pathway intersects with multiple cell signaling pathways, including the PI3K/AKT/mTOR pathway. Created with Biorender.com.
Figure 2 (a) Summary oncology treatment timeline. (b) Ca125 trend from diagnosis to current.
curroncol-28-00064-t001_Table 1Table 1 Food and Drug Administration approved indications for bevacizumab [18].
Cancer Stage Usage
Colorectal Metastatic, first-line 5 mg/kg every two weeks with bolus IFL
10 mg/kg every two weeks with FOLFOX4
Metastatic, recurrent after first-line bevacizumab-containing regimen 5 mg/kg every two weeks, or 7.5 mg/kg every three weeks with fluoropyrimidine–irinotecan, or fluoropyrimidine–oxaliplatin-based chemotherapy
Non-squamous, non-small-cell lung Unresectable, locally advanced, recurrent, or metastatic 15 mg/kg every three weeks with carboplatin and paclitaxel
Glioblastoma Recurrent 10 mg/kg every two weeks
Renal cell Metastatic 10 mg/kg every two weeks with interferon-alfa
Cervical Persistent, recurrent, or metastatic 15 mg/kg every three weeks with paclitaxel and cisplatin, or paclitaxel and topotecan
Epithelial ovarian, fallopian tube, or primary peritoneal III or IV, following surgical resection 15 mg/kg every three weeks with carboplatin and paclitaxel for up to six cycles, followed by 15 mg/kg every three weeks as a single agent for up to 22 cycles
Recurrent, platinum-sensitive 15 mg/kg every three weeks with carboplatin and either paclitaxel (6–8 cycles) or gemcitabine (6–10 cycles) followed by 15 mg/kg every 3 weeks as a single agent
Recurrent, platinum-resistant 10 mg/kg every two weeks with paclitaxel, pegylated liposomal doxorubicin, or topotecan given weekly
15 mg/kg every three weeks with topotecan every three weeks
Hepatocellular Unresectable or metastatic, first-line 15 mg/kg with atezolizumab every three weeks
Abbreviations: mg/kg = milligrams per kilogram; IFL = infusional fluoropyrimidine; FOLFOX4 = 5-fluorouracil, folic acid, and oxaliplatin.
curroncol-28-00064-t002_Table 2Table 2 Drug–drug interactions between post-transplant immunosuppressive medications and angiogenesis inhibitors.
Transplant Medication Potential Interactions with Anti-Angiogenesis Agents [26,27,28,29,30,31,32,33,34]
Cyclosporine
Increased cyclosporine levels and subsequent toxicity due to CYP3A4 and P-gp-mediated drug interactions (e.g., cabozatinib, axitinib, pazopanib, sorafenib, sunitinib)
Tacrolimus
Increased tacrolimus levels and subsequent toxicity due to inhibition of or competition with CYP3A4 metabolism and P-gp-mediated transport (e.g., cabozatinib, axitinib, pazopanib, sorafenib, sunitinib)
Additive impairment of the renal function (e.g., cediranib, axitinib, pazopanib)
Increased risk of QT prolongation with other agents that prolong the QT interval (e.g., cabozantinib, pazopanib, sorafenib, sunitinib)
Mycophenolate mofetil
Exaggerated leukopenia (e.g., ramucirumab, bevacizumab, sunitinib)
Azathioprine
Exaggerated leukopenia (e.g., ramucirumab, bevacizumab, sunitinib)
Sirolimus
Increased tacrolimus levels and subsequent toxicity due to inhibition of or competition with CYP3A4 metabolism and P-gp-mediated transport (e.g., cabozatinib, axitinib, pazopanib, sorafenib, sunitinib)
Additive impairment of the renal function (e.g., cediranib, axitinib, pazopanib)
Additive impairment of wound healing
Everolimus
Increased everolimus levels and subsequent toxicity due to inhibition of CYP3A4 metabolism and P-gp-mediated transport (e.g., cabozantinib, pazopanib)
Additive impairment of the renal function (e.g., cediranib, axitinib, pazopanib)
Additive impairment of wound healing
Corticosteroids
Competitive CYP3A4 metabolism (e.g., prednisone) with other CYP3A4 substrates (e.g., cabozantinib, axitinib, pazopanib, sorafenib, sunitinib)
Abbreviations: CYP3A4 = cytochrome P450 3A4; P-gp = P-glycoprotein. | Respiratory (inhalation) | DrugAdministrationRoute | CC BY | 33499164 | 18,991,561 | 2021-01-22 |
What was the dosage of drug 'ALBUTEROL'? | Angiogenesis Inhibitors as Anti-Cancer Therapy Following Renal Transplantation: A Case Report and Review of the Literature.
Solid organ transplant recipients on long-term immunosuppressive medication are at increased risk of developing malignancy, and treatment of advanced cancers with angiogenesis inhibitors in this context has not been widely studied. We present a case of recurrent high-grade serous ovarian carcinoma treated with paclitaxel and bevacizumab in the context of prior renal transplantation where the patient responded well to treatment with controlled toxicities, discussing the potential for increased rates of adverse events and drug interactions in this select population.
1. Introduction
Angiogenesis inhibitors such as vascular endothelial growth factor (VEGF) monoclonal antibodies and tyrosine kinase inhibitors (Figure 1) are standard treatments across various cancer subtypes. In advanced high-grade serous ovarian cancer (HGSOC), bevacizumab is part of the standard of care as maintenance therapy in front-line and recurrent disease [1]. Although the toxicity profile is generally well-managed, nephrotoxicity manifesting as proteinuria remains an important adverse event that requires close monitoring [2]. There is limited literature surrounding angiogenesis inhibitors as anti-cancer treatment in patients who have received solid organ transplants, and thus this report presents a case demonstrating safety of angiogenesis inhibition as anti-cancer therapy in a patient with a stable renal transplant.
2. Case Description
Consent: Fully informed, voluntary, written consent has been obtained to include patient information and publish this report.
Ethics: As this is a case report with fewer than three patients, institutional approval was not sought as per the University Health Network Research Ethics Board guidance document on case reports.
A 47-year-old woman presented with several months of dyspnea and abdominal distension to Princess Margaret Cancer Centre in February 2019 (Figure 2). CT revealed a 13.9 cm pelvic mass with peritoneal carcinomatosis and ascites. Omental biopsy confirmed HGSOC, and she received neoadjuvant platinum-based chemotherapy with excellent tolerance and no renal complications before proceeding to interval debulking in June 2019. There was no visible residual disease, and diagnosis of HGSOC was confirmed, germline and somatic BRCA wild type. She completed three cycles of adjuvant chemotherapy.
Her background was significant for IgA nephropathy, which resulted in progressive chronic kidney disease for 18 years prior to living donor kidney transplant in 2016. Both she and her donor were cytomegalovirus-positive, and she developed cytomegalovirus-associated colitis shortly post-transplantation. Her initial immunosuppression consisted of basiliximab induction followed by tacrolimus, mycophenolic acid, and steroids. She developed antibody-mediated rejection one week post-transplant, which was treated with plasmapheresis, immunoglobulin, and an increase in steroid dose. This was repeated three months later due to biopsy confirming ongoing antibody-mediated rejection. Mycophenolic acid was stopped upon HGSOC diagnosis. She has remained medication-adherent, with regular serum tacrolimus levels within the target range (most recently, 5.3 micrograms/L) and no signs of chronic graft rejection.
Her other comorbidities include diet-controlled, steroid-induced diabetes mellitus; ductal breast carcinoma in situ requiring wide local excision in 2011; asthma; reflux disease; hypertension; and hyperlipidemia. Other medications include prednisone 5 mg daily, acetylsalicylic acid, bisoprolol, trimethoprim–sulfamethoxazole, vitamin D, and inhaled salbutamol as needed.
In December 2019, she developed recurrence in the peritoneum and retroperitoneal lymph nodes, signifying platinum resistance. In January 2020, she commenced weekly paclitaxel 80 mg/m2 with bevacizumab 10 mg/kg every two weeks, and at that time, tacrolimus dose was reduced to 2 mg daily to aim for a serum level of 5 micrograms/L. She continued this therapy for over 6 months, and continued on the same dose of immunosuppression throughout with tacrolimus levels ranging between 3 and 7.6 micrograms/L. Serial imaging and Ca125 confirmed good response to treatment with reduction in size of tumor deposits. Her albumin/creatinine ratio was normal at 0.9 prior to the diagnosis of ovarian cancer, and most recently has been 15.6, signifying microalbuminuria; this has been monitored via urinalysis, which has consistently reported protein as negative or trace. Her most recent estimated glomerular filtration rate was 55 mL/min, similar to pre-diagnosis, and creatinine levels have mostly fluctuated between 95 and 120 umol/L.
Her course has been complicated by grade 2 hypertension (up to 145/95 mmHg) and non-cardiac chest pain, for which amlodipine was switched to ramipril 10 mg daily, and bevacizumab was withheld on two different occasions. Furthermore, she developed a brief period of Kidney Disease: Improving Global Outcomes (KDIGO) stage 1 acute kidney injury (AKI) of pre-renal etiology in September 2020 with creatinine of 160 umol/L, which self-resolved following withdrawal of bevacizumab, and was resumed after a 1-month-long break with no further episodes of kidney injury.
3. Discussion
Nephrotoxicity with angiogenesis inhibitors is relatively common, with proteinuria occurring in over 60% of patients [2]. Most cases are low-grade, transient, and do not require interventions or dose delays; however, more persistent, severe cases presenting as AKI and nephrotic syndrome can occur [2,3]. Risk factors associated with high-grade proteinuria include increased dose, prolonged administration, pre-existing renal disease, and administration of concurrent chemotherapy [4,5].
The pathophysiology of VEGF inhibitor-induced proteinuria remains unclear. Within a normal kidney, VEGF is produced by podocytes, and VEGF receptors are typically present on the glomerular and peritubular endothelium in addition to mesangial cells [6]. Inhibition of VEGF is thought to cause loss of endothelial fenestrations, podocyte injury and reduce endothelial proliferation, ultimately causing disruption of glomerular membranes [6]. Some cases have also demonstrated subacute thrombotic microangiopathy with endotheliosis and membranoproliferative changes [7]. Another manifestation of nephrotoxicity that is commonly seen is hypertension, occurring in more than a third of patients, which arises due to various mechanisms of renal vascular injury including inhibition of nitric oxide, rarefaction of microvasculature, and neuroendocrine dysregulation [8,9]. It is also hypothesized to increase intraglomerular pressure and ultrafiltration, leading to proteinuria [10].
The lack of clarity surrounding pathophysiology of VEGF inhibitor-induced nephrotoxicity [11] is demonstrated by the heterogeneity of published reports on glomerulopathy and other manifestations, including minimal change disease, collapsing glomerulopathy, membranoproliferative glomerulonephritis, focal segmental glomerulosclerosis, cryoglobulinemic glomerulonephritis, acute tubular necrosis, and interstitial nephritis [7]. Furthermore, worsening kidney disease can further exacerbate hypertension, which may perpetuate AKI [10].
Treatment for low-grade proteinuria usually includes an angiotensin-converting enzyme (ACE) inhibitor or an angiotensin receptor blocker to reduce glomerular filtration pressure, and anti-VEGF treatment can be continued providing stable proteinuria. However, proteinuria may worsen to the nephrotic range (>3 g in 24 h) with nephrotic syndrome. Even after cessation of VEGF inhibitor therapy, there are documented cases of persistent proteinuria [12].
3.1. Angiogenesis Inhibitors Post-Transplant
The role of the VEGF pathway in the pathogenesis of post-transplant complications is poorly understood, with preliminary reports across various organ transplants showing hypothesis-generating results [13,14,15,16]. Upregulation of angiogenesis factors was associated with increased allograft vasculopathy, bronchiolitis obliterans, and recurrence of hepatocellular carcinoma in cardiac, pulmonary, and liver transplants, respectively [13,15,16,17]. In renal transplants, VEGF is thought to be upregulated in acute and chronic rejection, particularly associated with cyclosporine [14].
As the first VEGF inhibitor used for anti-cancer treatment, bevacizumab was approved by the United States Food and Drug Administration in 2004, and is now licensed for use in many cancers [18] (Table 1). In ovarian cancer, patients who were prior recipients of solid organ transplants or who were receiving immunosuppressive therapies were not excluded from randomized bevacizumab trials, but those with pre-existing uncontrolled hypertension or renal dysfunction based on serum creatinine ≥ 1.6 mg/dL or proteinuria > 1 g per 24 h were excluded [19,20,21]. Similarly, in other large randomized angiogenesis inhibitor studies across other tumor sites, prior solid organ transplant or use of immunosuppressants is not an exclusion criterion, aside from studies involving immune checkpoint inhibitors [22]. Identification of patients enrolled in large angiogenesis inhibitor trials who had received prior transplants could potentially make for an interesting post-hoc pooled analysis.
Similarly, reports on angiogenesis inhibition in solid organ transplant patients remain scant in the literature, as highlighted by a review on bevacizumab toxicity by Fenoglio et al [9]. Musri et al. reported a case of colorectal cancer post-renal transplantation with baseline proteinuria, which significantly worsened on administration of intravenous 5-fluorouracil, irinotecan, oxaliplatin, and bevacizumab [23]. Cheungpasitporn et al. described two cases with renal allograft dysfunction following administration of intravitreal bevacizumab, aflibercept, or ranibizumab [24]. Doses of anti-angiogenics were lower but not specified within this report. Although neither case proved causality with anti-VEGF therapy, one was diagnosed with phospholipase A2 receptor-negative membranous nephropathy, and the second revealed acute and chronic antibody-mediated rejection with glomerular thrombi and transplant glomerulopathy. Jonkers and Buren reported a case of worsening IgA nephropathy presenting with nephrotic-range proteinuria post-renal transplantation on sorafenib [25]. These reports highlight the potential severe nephrotoxicity known to be associated with angiogenesis inhibitors; however, there remain few documented positive experiences with angiogenesis inhibitor use in the post-transplant setting. Given the prevalence of nephrotoxicity with these agents, these considerations are particularly prudent for renal transplant recipients, but reports in other organ transplants remain similarly scarce.
3.2. Medication Interactions
Solid organ transplant recipients frequently take maintenance immunosuppressive agents, including but not limited to corticosteroids, calcineurin inhibitors, anti-proliferative agents, and mTOR inhibitors, which are associated with various complications and drug interactions (Table 2).
In the case presented, tacrolimus levels were measured every few months to be within the therapeutic range; this is significant as potential drug interactions between transplant medications and angiogenesis inhibitors involve pharmacokinetic and pharmacodynamic interactions. Pharmacokinetic interactions typically occur due to cytochrome P450 enzyme (CYP) and P-glycoprotein (P-gp) drug transport systems [26,27,28], and risk of competitive metabolism as substrates for the same enzyme or transporter may increase serum levels. Axitinib and sorafenib are CYP3A4 and P-gp substrates, and cabozantinib and pazopanib are substrates and inhibitors of both enzyme systems [29,30,31,32]. Drug interactions are well-described for CYP3A4 substrates cyclosporine, tacrolimus, and sirolimus, with metabolic inhibition leading to increased immunosuppressant concentrations (~20%) [28]. Pharmacodynamic interactions primarily concern cumulative toxicities between these two classes (Table 2) [26,27,33,34]. Interestingly, Onodera et al. reported upon a case of metastatic colorectal cancer post-renal transplant where a patient was administered five cycles of 5-fluorouracil, oxaliplatin, and bevacizumab where although severe proteinuria occurred, serum tacrolimus levels were not affected throughout the course of treatment [35]. This report remains one of the only cases in the literature that demonstrates stability of immunosuppression whilst on bevacizumab post-transplant.
3.3. Long-Term Adverse Events
One of the leading causes of morbidity and mortality in solid organ transplant recipients is malignancy, most commonly non-melanomatous skin cancers [36]. Other malignancies such as colorectal, kidney, and cervical cancers are also prevalent in the post-transplant context, and angiogenesis inhibitors such as bevacizumab are commonly used in metastatic disease [18] (Table 1). Surveillance recommendations within transplant recipients are variable across the globe due to a paucity of robust screening trials [37]. Other long-term complications associated with organ transplantation and prolonged immunosuppressant use include cardiovascular disease, diabetes mellitus, hypertension, and infection associated with cytopenia. In the patient presented, adverse events have not outweighed benefits of ongoing treatment, but this will need close monitoring given the risk of overlapping toxicities as long-term adverse event data remain limited [3,38].
4. Conclusions
Whilst there are minimal data justifying that bevacizumab or other angiogenesis inhibitors are unsafe in the post-transplant setting, there is similarly scarce literature demonstrating safe administration, as in the patient presented. As life expectancy continues to improve with increasing indications for transplantation, long-term risks for malignancy with prolonged immunosuppression are increasingly relevant as a cause of mortality in solid organ transplant recipients. In patients who have undergone renal transplantation, careful consideration of treatment options with risk of nephrotoxicity and close monitoring remains paramount.
Although treating oncologists should remain vigilant about potential drug interactions and overlapping toxicities, these are not necessarily contraindications for agents such as bevacizumab. Treatment decisions should consider the best available evidence, and collating information about toxicity and tolerance from randomized trials and post-approval Phase IV studies would provide detailed information from at-risk subgroups. This calls for a stratified, inclusive approach to allow enrolment of those with chronic diseases and comorbidities in prospective trials, allowing objective assessment of the risk–benefit ratio.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Author Contributions
Conceptualization, L.K., A.M.O.; methodology, L.K., J.D.; software, L.K.; validation, L.K., J.D., J.S., A.K.; resources, L.K.; data curation, L.K., S.L., J.S.; writing—original draft preparation, L.K., J.D.; writing—review and editing, L.K., J.D., K.K., A.K., J.S., S.L., A.M.O.; visualization, L.K., K.K.; supervision, S.L., A.M.O. All authors have read and agreed to the published version of the manuscript.
Funding
This research received no external funding.
Institutional Review Board Statement
As this was a report with fewer than three patients, institutional approval was not sought as per the University Health Network Research Ethics Board guidance document on case reports.
Informed Consent Statement
Informed consent was obtained from all subjects involved in the study and written informed consent has been obtained from the patient to publish this paper.
Data Availability Statement
Not applicable.
Conflicts of Interest
We have read and understood Current Oncology’s policy on disclosing conflicts of interest and declare that we have none.
Abbreviations
VEGFR vascular endothelial growth factor receptor
EGFR epidermal growth factor receptor
FGFR fibroblast growth factor receptor
PDGFR platelet-derived growth factor receptor
SCF stem cell factor
PI3K phosphoinositide 3-kinase
AKT protein kinase B
mTOR mammalian target of rapamycin
RET RET proto-oncogene
AXL AXL receptor tyrosine kinase
TAH–BSO total abdominal hysterectomy and bilateral salpingo-oophorectomy
Ca125 cancer antigen 125
U/ml units per milliliter
AKI acute kidney injury
CYP cytochrome P450
HGSOC high-grade serous ovarian carcinoma
FOLFOX4 5-fluorouracil, leucovorin, and oxaliplatin
IFL irinotecan, leucovorin, and 5-fluorouracil
P-gp P-glycoprotein
Figure 1 Angiogenesis pathways in malignancy and targeted therapies commonly used. Monoclonal antibodies including bevacizumab and ramucirumab and tyrosine kinase inhibitors including sunitinib, sorafenib, lenvatinib, pazopanib, axitinib, cediranib, and cabozantinib have been included (this list is not exhaustive). Receptors including VEGFR, c-kit, epidermal growth factor receptor (EGFR), fibroblast growth factor receptor (FGFR), and PDGFR are displayed here, but certain medications also target other pathways not displayed here such as AXL and RET. The VEGFR pathway intersects with multiple cell signaling pathways, including the PI3K/AKT/mTOR pathway. Created with Biorender.com.
Figure 2 (a) Summary oncology treatment timeline. (b) Ca125 trend from diagnosis to current.
curroncol-28-00064-t001_Table 1Table 1 Food and Drug Administration approved indications for bevacizumab [18].
Cancer Stage Usage
Colorectal Metastatic, first-line 5 mg/kg every two weeks with bolus IFL
10 mg/kg every two weeks with FOLFOX4
Metastatic, recurrent after first-line bevacizumab-containing regimen 5 mg/kg every two weeks, or 7.5 mg/kg every three weeks with fluoropyrimidine–irinotecan, or fluoropyrimidine–oxaliplatin-based chemotherapy
Non-squamous, non-small-cell lung Unresectable, locally advanced, recurrent, or metastatic 15 mg/kg every three weeks with carboplatin and paclitaxel
Glioblastoma Recurrent 10 mg/kg every two weeks
Renal cell Metastatic 10 mg/kg every two weeks with interferon-alfa
Cervical Persistent, recurrent, or metastatic 15 mg/kg every three weeks with paclitaxel and cisplatin, or paclitaxel and topotecan
Epithelial ovarian, fallopian tube, or primary peritoneal III or IV, following surgical resection 15 mg/kg every three weeks with carboplatin and paclitaxel for up to six cycles, followed by 15 mg/kg every three weeks as a single agent for up to 22 cycles
Recurrent, platinum-sensitive 15 mg/kg every three weeks with carboplatin and either paclitaxel (6–8 cycles) or gemcitabine (6–10 cycles) followed by 15 mg/kg every 3 weeks as a single agent
Recurrent, platinum-resistant 10 mg/kg every two weeks with paclitaxel, pegylated liposomal doxorubicin, or topotecan given weekly
15 mg/kg every three weeks with topotecan every three weeks
Hepatocellular Unresectable or metastatic, first-line 15 mg/kg with atezolizumab every three weeks
Abbreviations: mg/kg = milligrams per kilogram; IFL = infusional fluoropyrimidine; FOLFOX4 = 5-fluorouracil, folic acid, and oxaliplatin.
curroncol-28-00064-t002_Table 2Table 2 Drug–drug interactions between post-transplant immunosuppressive medications and angiogenesis inhibitors.
Transplant Medication Potential Interactions with Anti-Angiogenesis Agents [26,27,28,29,30,31,32,33,34]
Cyclosporine
Increased cyclosporine levels and subsequent toxicity due to CYP3A4 and P-gp-mediated drug interactions (e.g., cabozatinib, axitinib, pazopanib, sorafenib, sunitinib)
Tacrolimus
Increased tacrolimus levels and subsequent toxicity due to inhibition of or competition with CYP3A4 metabolism and P-gp-mediated transport (e.g., cabozatinib, axitinib, pazopanib, sorafenib, sunitinib)
Additive impairment of the renal function (e.g., cediranib, axitinib, pazopanib)
Increased risk of QT prolongation with other agents that prolong the QT interval (e.g., cabozantinib, pazopanib, sorafenib, sunitinib)
Mycophenolate mofetil
Exaggerated leukopenia (e.g., ramucirumab, bevacizumab, sunitinib)
Azathioprine
Exaggerated leukopenia (e.g., ramucirumab, bevacizumab, sunitinib)
Sirolimus
Increased tacrolimus levels and subsequent toxicity due to inhibition of or competition with CYP3A4 metabolism and P-gp-mediated transport (e.g., cabozatinib, axitinib, pazopanib, sorafenib, sunitinib)
Additive impairment of the renal function (e.g., cediranib, axitinib, pazopanib)
Additive impairment of wound healing
Everolimus
Increased everolimus levels and subsequent toxicity due to inhibition of CYP3A4 metabolism and P-gp-mediated transport (e.g., cabozantinib, pazopanib)
Additive impairment of the renal function (e.g., cediranib, axitinib, pazopanib)
Additive impairment of wound healing
Corticosteroids
Competitive CYP3A4 metabolism (e.g., prednisone) with other CYP3A4 substrates (e.g., cabozantinib, axitinib, pazopanib, sorafenib, sunitinib)
Abbreviations: CYP3A4 = cytochrome P450 3A4; P-gp = P-glycoprotein. | UNK, AS NEEDED | DrugDosageText | CC BY | 33499164 | 18,991,561 | 2021-01-22 |
What was the dosage of drug 'BEVACIZUMAB'? | Angiogenesis Inhibitors as Anti-Cancer Therapy Following Renal Transplantation: A Case Report and Review of the Literature.
Solid organ transplant recipients on long-term immunosuppressive medication are at increased risk of developing malignancy, and treatment of advanced cancers with angiogenesis inhibitors in this context has not been widely studied. We present a case of recurrent high-grade serous ovarian carcinoma treated with paclitaxel and bevacizumab in the context of prior renal transplantation where the patient responded well to treatment with controlled toxicities, discussing the potential for increased rates of adverse events and drug interactions in this select population.
1. Introduction
Angiogenesis inhibitors such as vascular endothelial growth factor (VEGF) monoclonal antibodies and tyrosine kinase inhibitors (Figure 1) are standard treatments across various cancer subtypes. In advanced high-grade serous ovarian cancer (HGSOC), bevacizumab is part of the standard of care as maintenance therapy in front-line and recurrent disease [1]. Although the toxicity profile is generally well-managed, nephrotoxicity manifesting as proteinuria remains an important adverse event that requires close monitoring [2]. There is limited literature surrounding angiogenesis inhibitors as anti-cancer treatment in patients who have received solid organ transplants, and thus this report presents a case demonstrating safety of angiogenesis inhibition as anti-cancer therapy in a patient with a stable renal transplant.
2. Case Description
Consent: Fully informed, voluntary, written consent has been obtained to include patient information and publish this report.
Ethics: As this is a case report with fewer than three patients, institutional approval was not sought as per the University Health Network Research Ethics Board guidance document on case reports.
A 47-year-old woman presented with several months of dyspnea and abdominal distension to Princess Margaret Cancer Centre in February 2019 (Figure 2). CT revealed a 13.9 cm pelvic mass with peritoneal carcinomatosis and ascites. Omental biopsy confirmed HGSOC, and she received neoadjuvant platinum-based chemotherapy with excellent tolerance and no renal complications before proceeding to interval debulking in June 2019. There was no visible residual disease, and diagnosis of HGSOC was confirmed, germline and somatic BRCA wild type. She completed three cycles of adjuvant chemotherapy.
Her background was significant for IgA nephropathy, which resulted in progressive chronic kidney disease for 18 years prior to living donor kidney transplant in 2016. Both she and her donor were cytomegalovirus-positive, and she developed cytomegalovirus-associated colitis shortly post-transplantation. Her initial immunosuppression consisted of basiliximab induction followed by tacrolimus, mycophenolic acid, and steroids. She developed antibody-mediated rejection one week post-transplant, which was treated with plasmapheresis, immunoglobulin, and an increase in steroid dose. This was repeated three months later due to biopsy confirming ongoing antibody-mediated rejection. Mycophenolic acid was stopped upon HGSOC diagnosis. She has remained medication-adherent, with regular serum tacrolimus levels within the target range (most recently, 5.3 micrograms/L) and no signs of chronic graft rejection.
Her other comorbidities include diet-controlled, steroid-induced diabetes mellitus; ductal breast carcinoma in situ requiring wide local excision in 2011; asthma; reflux disease; hypertension; and hyperlipidemia. Other medications include prednisone 5 mg daily, acetylsalicylic acid, bisoprolol, trimethoprim–sulfamethoxazole, vitamin D, and inhaled salbutamol as needed.
In December 2019, she developed recurrence in the peritoneum and retroperitoneal lymph nodes, signifying platinum resistance. In January 2020, she commenced weekly paclitaxel 80 mg/m2 with bevacizumab 10 mg/kg every two weeks, and at that time, tacrolimus dose was reduced to 2 mg daily to aim for a serum level of 5 micrograms/L. She continued this therapy for over 6 months, and continued on the same dose of immunosuppression throughout with tacrolimus levels ranging between 3 and 7.6 micrograms/L. Serial imaging and Ca125 confirmed good response to treatment with reduction in size of tumor deposits. Her albumin/creatinine ratio was normal at 0.9 prior to the diagnosis of ovarian cancer, and most recently has been 15.6, signifying microalbuminuria; this has been monitored via urinalysis, which has consistently reported protein as negative or trace. Her most recent estimated glomerular filtration rate was 55 mL/min, similar to pre-diagnosis, and creatinine levels have mostly fluctuated between 95 and 120 umol/L.
Her course has been complicated by grade 2 hypertension (up to 145/95 mmHg) and non-cardiac chest pain, for which amlodipine was switched to ramipril 10 mg daily, and bevacizumab was withheld on two different occasions. Furthermore, she developed a brief period of Kidney Disease: Improving Global Outcomes (KDIGO) stage 1 acute kidney injury (AKI) of pre-renal etiology in September 2020 with creatinine of 160 umol/L, which self-resolved following withdrawal of bevacizumab, and was resumed after a 1-month-long break with no further episodes of kidney injury.
3. Discussion
Nephrotoxicity with angiogenesis inhibitors is relatively common, with proteinuria occurring in over 60% of patients [2]. Most cases are low-grade, transient, and do not require interventions or dose delays; however, more persistent, severe cases presenting as AKI and nephrotic syndrome can occur [2,3]. Risk factors associated with high-grade proteinuria include increased dose, prolonged administration, pre-existing renal disease, and administration of concurrent chemotherapy [4,5].
The pathophysiology of VEGF inhibitor-induced proteinuria remains unclear. Within a normal kidney, VEGF is produced by podocytes, and VEGF receptors are typically present on the glomerular and peritubular endothelium in addition to mesangial cells [6]. Inhibition of VEGF is thought to cause loss of endothelial fenestrations, podocyte injury and reduce endothelial proliferation, ultimately causing disruption of glomerular membranes [6]. Some cases have also demonstrated subacute thrombotic microangiopathy with endotheliosis and membranoproliferative changes [7]. Another manifestation of nephrotoxicity that is commonly seen is hypertension, occurring in more than a third of patients, which arises due to various mechanisms of renal vascular injury including inhibition of nitric oxide, rarefaction of microvasculature, and neuroendocrine dysregulation [8,9]. It is also hypothesized to increase intraglomerular pressure and ultrafiltration, leading to proteinuria [10].
The lack of clarity surrounding pathophysiology of VEGF inhibitor-induced nephrotoxicity [11] is demonstrated by the heterogeneity of published reports on glomerulopathy and other manifestations, including minimal change disease, collapsing glomerulopathy, membranoproliferative glomerulonephritis, focal segmental glomerulosclerosis, cryoglobulinemic glomerulonephritis, acute tubular necrosis, and interstitial nephritis [7]. Furthermore, worsening kidney disease can further exacerbate hypertension, which may perpetuate AKI [10].
Treatment for low-grade proteinuria usually includes an angiotensin-converting enzyme (ACE) inhibitor or an angiotensin receptor blocker to reduce glomerular filtration pressure, and anti-VEGF treatment can be continued providing stable proteinuria. However, proteinuria may worsen to the nephrotic range (>3 g in 24 h) with nephrotic syndrome. Even after cessation of VEGF inhibitor therapy, there are documented cases of persistent proteinuria [12].
3.1. Angiogenesis Inhibitors Post-Transplant
The role of the VEGF pathway in the pathogenesis of post-transplant complications is poorly understood, with preliminary reports across various organ transplants showing hypothesis-generating results [13,14,15,16]. Upregulation of angiogenesis factors was associated with increased allograft vasculopathy, bronchiolitis obliterans, and recurrence of hepatocellular carcinoma in cardiac, pulmonary, and liver transplants, respectively [13,15,16,17]. In renal transplants, VEGF is thought to be upregulated in acute and chronic rejection, particularly associated with cyclosporine [14].
As the first VEGF inhibitor used for anti-cancer treatment, bevacizumab was approved by the United States Food and Drug Administration in 2004, and is now licensed for use in many cancers [18] (Table 1). In ovarian cancer, patients who were prior recipients of solid organ transplants or who were receiving immunosuppressive therapies were not excluded from randomized bevacizumab trials, but those with pre-existing uncontrolled hypertension or renal dysfunction based on serum creatinine ≥ 1.6 mg/dL or proteinuria > 1 g per 24 h were excluded [19,20,21]. Similarly, in other large randomized angiogenesis inhibitor studies across other tumor sites, prior solid organ transplant or use of immunosuppressants is not an exclusion criterion, aside from studies involving immune checkpoint inhibitors [22]. Identification of patients enrolled in large angiogenesis inhibitor trials who had received prior transplants could potentially make for an interesting post-hoc pooled analysis.
Similarly, reports on angiogenesis inhibition in solid organ transplant patients remain scant in the literature, as highlighted by a review on bevacizumab toxicity by Fenoglio et al [9]. Musri et al. reported a case of colorectal cancer post-renal transplantation with baseline proteinuria, which significantly worsened on administration of intravenous 5-fluorouracil, irinotecan, oxaliplatin, and bevacizumab [23]. Cheungpasitporn et al. described two cases with renal allograft dysfunction following administration of intravitreal bevacizumab, aflibercept, or ranibizumab [24]. Doses of anti-angiogenics were lower but not specified within this report. Although neither case proved causality with anti-VEGF therapy, one was diagnosed with phospholipase A2 receptor-negative membranous nephropathy, and the second revealed acute and chronic antibody-mediated rejection with glomerular thrombi and transplant glomerulopathy. Jonkers and Buren reported a case of worsening IgA nephropathy presenting with nephrotic-range proteinuria post-renal transplantation on sorafenib [25]. These reports highlight the potential severe nephrotoxicity known to be associated with angiogenesis inhibitors; however, there remain few documented positive experiences with angiogenesis inhibitor use in the post-transplant setting. Given the prevalence of nephrotoxicity with these agents, these considerations are particularly prudent for renal transplant recipients, but reports in other organ transplants remain similarly scarce.
3.2. Medication Interactions
Solid organ transplant recipients frequently take maintenance immunosuppressive agents, including but not limited to corticosteroids, calcineurin inhibitors, anti-proliferative agents, and mTOR inhibitors, which are associated with various complications and drug interactions (Table 2).
In the case presented, tacrolimus levels were measured every few months to be within the therapeutic range; this is significant as potential drug interactions between transplant medications and angiogenesis inhibitors involve pharmacokinetic and pharmacodynamic interactions. Pharmacokinetic interactions typically occur due to cytochrome P450 enzyme (CYP) and P-glycoprotein (P-gp) drug transport systems [26,27,28], and risk of competitive metabolism as substrates for the same enzyme or transporter may increase serum levels. Axitinib and sorafenib are CYP3A4 and P-gp substrates, and cabozantinib and pazopanib are substrates and inhibitors of both enzyme systems [29,30,31,32]. Drug interactions are well-described for CYP3A4 substrates cyclosporine, tacrolimus, and sirolimus, with metabolic inhibition leading to increased immunosuppressant concentrations (~20%) [28]. Pharmacodynamic interactions primarily concern cumulative toxicities between these two classes (Table 2) [26,27,33,34]. Interestingly, Onodera et al. reported upon a case of metastatic colorectal cancer post-renal transplant where a patient was administered five cycles of 5-fluorouracil, oxaliplatin, and bevacizumab where although severe proteinuria occurred, serum tacrolimus levels were not affected throughout the course of treatment [35]. This report remains one of the only cases in the literature that demonstrates stability of immunosuppression whilst on bevacizumab post-transplant.
3.3. Long-Term Adverse Events
One of the leading causes of morbidity and mortality in solid organ transplant recipients is malignancy, most commonly non-melanomatous skin cancers [36]. Other malignancies such as colorectal, kidney, and cervical cancers are also prevalent in the post-transplant context, and angiogenesis inhibitors such as bevacizumab are commonly used in metastatic disease [18] (Table 1). Surveillance recommendations within transplant recipients are variable across the globe due to a paucity of robust screening trials [37]. Other long-term complications associated with organ transplantation and prolonged immunosuppressant use include cardiovascular disease, diabetes mellitus, hypertension, and infection associated with cytopenia. In the patient presented, adverse events have not outweighed benefits of ongoing treatment, but this will need close monitoring given the risk of overlapping toxicities as long-term adverse event data remain limited [3,38].
4. Conclusions
Whilst there are minimal data justifying that bevacizumab or other angiogenesis inhibitors are unsafe in the post-transplant setting, there is similarly scarce literature demonstrating safe administration, as in the patient presented. As life expectancy continues to improve with increasing indications for transplantation, long-term risks for malignancy with prolonged immunosuppression are increasingly relevant as a cause of mortality in solid organ transplant recipients. In patients who have undergone renal transplantation, careful consideration of treatment options with risk of nephrotoxicity and close monitoring remains paramount.
Although treating oncologists should remain vigilant about potential drug interactions and overlapping toxicities, these are not necessarily contraindications for agents such as bevacizumab. Treatment decisions should consider the best available evidence, and collating information about toxicity and tolerance from randomized trials and post-approval Phase IV studies would provide detailed information from at-risk subgroups. This calls for a stratified, inclusive approach to allow enrolment of those with chronic diseases and comorbidities in prospective trials, allowing objective assessment of the risk–benefit ratio.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Author Contributions
Conceptualization, L.K., A.M.O.; methodology, L.K., J.D.; software, L.K.; validation, L.K., J.D., J.S., A.K.; resources, L.K.; data curation, L.K., S.L., J.S.; writing—original draft preparation, L.K., J.D.; writing—review and editing, L.K., J.D., K.K., A.K., J.S., S.L., A.M.O.; visualization, L.K., K.K.; supervision, S.L., A.M.O. All authors have read and agreed to the published version of the manuscript.
Funding
This research received no external funding.
Institutional Review Board Statement
As this was a report with fewer than three patients, institutional approval was not sought as per the University Health Network Research Ethics Board guidance document on case reports.
Informed Consent Statement
Informed consent was obtained from all subjects involved in the study and written informed consent has been obtained from the patient to publish this paper.
Data Availability Statement
Not applicable.
Conflicts of Interest
We have read and understood Current Oncology’s policy on disclosing conflicts of interest and declare that we have none.
Abbreviations
VEGFR vascular endothelial growth factor receptor
EGFR epidermal growth factor receptor
FGFR fibroblast growth factor receptor
PDGFR platelet-derived growth factor receptor
SCF stem cell factor
PI3K phosphoinositide 3-kinase
AKT protein kinase B
mTOR mammalian target of rapamycin
RET RET proto-oncogene
AXL AXL receptor tyrosine kinase
TAH–BSO total abdominal hysterectomy and bilateral salpingo-oophorectomy
Ca125 cancer antigen 125
U/ml units per milliliter
AKI acute kidney injury
CYP cytochrome P450
HGSOC high-grade serous ovarian carcinoma
FOLFOX4 5-fluorouracil, leucovorin, and oxaliplatin
IFL irinotecan, leucovorin, and 5-fluorouracil
P-gp P-glycoprotein
Figure 1 Angiogenesis pathways in malignancy and targeted therapies commonly used. Monoclonal antibodies including bevacizumab and ramucirumab and tyrosine kinase inhibitors including sunitinib, sorafenib, lenvatinib, pazopanib, axitinib, cediranib, and cabozantinib have been included (this list is not exhaustive). Receptors including VEGFR, c-kit, epidermal growth factor receptor (EGFR), fibroblast growth factor receptor (FGFR), and PDGFR are displayed here, but certain medications also target other pathways not displayed here such as AXL and RET. The VEGFR pathway intersects with multiple cell signaling pathways, including the PI3K/AKT/mTOR pathway. Created with Biorender.com.
Figure 2 (a) Summary oncology treatment timeline. (b) Ca125 trend from diagnosis to current.
curroncol-28-00064-t001_Table 1Table 1 Food and Drug Administration approved indications for bevacizumab [18].
Cancer Stage Usage
Colorectal Metastatic, first-line 5 mg/kg every two weeks with bolus IFL
10 mg/kg every two weeks with FOLFOX4
Metastatic, recurrent after first-line bevacizumab-containing regimen 5 mg/kg every two weeks, or 7.5 mg/kg every three weeks with fluoropyrimidine–irinotecan, or fluoropyrimidine–oxaliplatin-based chemotherapy
Non-squamous, non-small-cell lung Unresectable, locally advanced, recurrent, or metastatic 15 mg/kg every three weeks with carboplatin and paclitaxel
Glioblastoma Recurrent 10 mg/kg every two weeks
Renal cell Metastatic 10 mg/kg every two weeks with interferon-alfa
Cervical Persistent, recurrent, or metastatic 15 mg/kg every three weeks with paclitaxel and cisplatin, or paclitaxel and topotecan
Epithelial ovarian, fallopian tube, or primary peritoneal III or IV, following surgical resection 15 mg/kg every three weeks with carboplatin and paclitaxel for up to six cycles, followed by 15 mg/kg every three weeks as a single agent for up to 22 cycles
Recurrent, platinum-sensitive 15 mg/kg every three weeks with carboplatin and either paclitaxel (6–8 cycles) or gemcitabine (6–10 cycles) followed by 15 mg/kg every 3 weeks as a single agent
Recurrent, platinum-resistant 10 mg/kg every two weeks with paclitaxel, pegylated liposomal doxorubicin, or topotecan given weekly
15 mg/kg every three weeks with topotecan every three weeks
Hepatocellular Unresectable or metastatic, first-line 15 mg/kg with atezolizumab every three weeks
Abbreviations: mg/kg = milligrams per kilogram; IFL = infusional fluoropyrimidine; FOLFOX4 = 5-fluorouracil, folic acid, and oxaliplatin.
curroncol-28-00064-t002_Table 2Table 2 Drug–drug interactions between post-transplant immunosuppressive medications and angiogenesis inhibitors.
Transplant Medication Potential Interactions with Anti-Angiogenesis Agents [26,27,28,29,30,31,32,33,34]
Cyclosporine
Increased cyclosporine levels and subsequent toxicity due to CYP3A4 and P-gp-mediated drug interactions (e.g., cabozatinib, axitinib, pazopanib, sorafenib, sunitinib)
Tacrolimus
Increased tacrolimus levels and subsequent toxicity due to inhibition of or competition with CYP3A4 metabolism and P-gp-mediated transport (e.g., cabozatinib, axitinib, pazopanib, sorafenib, sunitinib)
Additive impairment of the renal function (e.g., cediranib, axitinib, pazopanib)
Increased risk of QT prolongation with other agents that prolong the QT interval (e.g., cabozantinib, pazopanib, sorafenib, sunitinib)
Mycophenolate mofetil
Exaggerated leukopenia (e.g., ramucirumab, bevacizumab, sunitinib)
Azathioprine
Exaggerated leukopenia (e.g., ramucirumab, bevacizumab, sunitinib)
Sirolimus
Increased tacrolimus levels and subsequent toxicity due to inhibition of or competition with CYP3A4 metabolism and P-gp-mediated transport (e.g., cabozatinib, axitinib, pazopanib, sorafenib, sunitinib)
Additive impairment of the renal function (e.g., cediranib, axitinib, pazopanib)
Additive impairment of wound healing
Everolimus
Increased everolimus levels and subsequent toxicity due to inhibition of CYP3A4 metabolism and P-gp-mediated transport (e.g., cabozantinib, pazopanib)
Additive impairment of the renal function (e.g., cediranib, axitinib, pazopanib)
Additive impairment of wound healing
Corticosteroids
Competitive CYP3A4 metabolism (e.g., prednisone) with other CYP3A4 substrates (e.g., cabozantinib, axitinib, pazopanib, sorafenib, sunitinib)
Abbreviations: CYP3A4 = cytochrome P450 3A4; P-gp = P-glycoprotein. | 10 MG/KG, CYCLIC (EVERY TWO WEEKS) | DrugDosageText | CC BY | 33499164 | 18,991,561 | 2021-01-22 |
What was the dosage of drug 'PREDNISONE'? | Angiogenesis Inhibitors as Anti-Cancer Therapy Following Renal Transplantation: A Case Report and Review of the Literature.
Solid organ transplant recipients on long-term immunosuppressive medication are at increased risk of developing malignancy, and treatment of advanced cancers with angiogenesis inhibitors in this context has not been widely studied. We present a case of recurrent high-grade serous ovarian carcinoma treated with paclitaxel and bevacizumab in the context of prior renal transplantation where the patient responded well to treatment with controlled toxicities, discussing the potential for increased rates of adverse events and drug interactions in this select population.
1. Introduction
Angiogenesis inhibitors such as vascular endothelial growth factor (VEGF) monoclonal antibodies and tyrosine kinase inhibitors (Figure 1) are standard treatments across various cancer subtypes. In advanced high-grade serous ovarian cancer (HGSOC), bevacizumab is part of the standard of care as maintenance therapy in front-line and recurrent disease [1]. Although the toxicity profile is generally well-managed, nephrotoxicity manifesting as proteinuria remains an important adverse event that requires close monitoring [2]. There is limited literature surrounding angiogenesis inhibitors as anti-cancer treatment in patients who have received solid organ transplants, and thus this report presents a case demonstrating safety of angiogenesis inhibition as anti-cancer therapy in a patient with a stable renal transplant.
2. Case Description
Consent: Fully informed, voluntary, written consent has been obtained to include patient information and publish this report.
Ethics: As this is a case report with fewer than three patients, institutional approval was not sought as per the University Health Network Research Ethics Board guidance document on case reports.
A 47-year-old woman presented with several months of dyspnea and abdominal distension to Princess Margaret Cancer Centre in February 2019 (Figure 2). CT revealed a 13.9 cm pelvic mass with peritoneal carcinomatosis and ascites. Omental biopsy confirmed HGSOC, and she received neoadjuvant platinum-based chemotherapy with excellent tolerance and no renal complications before proceeding to interval debulking in June 2019. There was no visible residual disease, and diagnosis of HGSOC was confirmed, germline and somatic BRCA wild type. She completed three cycles of adjuvant chemotherapy.
Her background was significant for IgA nephropathy, which resulted in progressive chronic kidney disease for 18 years prior to living donor kidney transplant in 2016. Both she and her donor were cytomegalovirus-positive, and she developed cytomegalovirus-associated colitis shortly post-transplantation. Her initial immunosuppression consisted of basiliximab induction followed by tacrolimus, mycophenolic acid, and steroids. She developed antibody-mediated rejection one week post-transplant, which was treated with plasmapheresis, immunoglobulin, and an increase in steroid dose. This was repeated three months later due to biopsy confirming ongoing antibody-mediated rejection. Mycophenolic acid was stopped upon HGSOC diagnosis. She has remained medication-adherent, with regular serum tacrolimus levels within the target range (most recently, 5.3 micrograms/L) and no signs of chronic graft rejection.
Her other comorbidities include diet-controlled, steroid-induced diabetes mellitus; ductal breast carcinoma in situ requiring wide local excision in 2011; asthma; reflux disease; hypertension; and hyperlipidemia. Other medications include prednisone 5 mg daily, acetylsalicylic acid, bisoprolol, trimethoprim–sulfamethoxazole, vitamin D, and inhaled salbutamol as needed.
In December 2019, she developed recurrence in the peritoneum and retroperitoneal lymph nodes, signifying platinum resistance. In January 2020, she commenced weekly paclitaxel 80 mg/m2 with bevacizumab 10 mg/kg every two weeks, and at that time, tacrolimus dose was reduced to 2 mg daily to aim for a serum level of 5 micrograms/L. She continued this therapy for over 6 months, and continued on the same dose of immunosuppression throughout with tacrolimus levels ranging between 3 and 7.6 micrograms/L. Serial imaging and Ca125 confirmed good response to treatment with reduction in size of tumor deposits. Her albumin/creatinine ratio was normal at 0.9 prior to the diagnosis of ovarian cancer, and most recently has been 15.6, signifying microalbuminuria; this has been monitored via urinalysis, which has consistently reported protein as negative or trace. Her most recent estimated glomerular filtration rate was 55 mL/min, similar to pre-diagnosis, and creatinine levels have mostly fluctuated between 95 and 120 umol/L.
Her course has been complicated by grade 2 hypertension (up to 145/95 mmHg) and non-cardiac chest pain, for which amlodipine was switched to ramipril 10 mg daily, and bevacizumab was withheld on two different occasions. Furthermore, she developed a brief period of Kidney Disease: Improving Global Outcomes (KDIGO) stage 1 acute kidney injury (AKI) of pre-renal etiology in September 2020 with creatinine of 160 umol/L, which self-resolved following withdrawal of bevacizumab, and was resumed after a 1-month-long break with no further episodes of kidney injury.
3. Discussion
Nephrotoxicity with angiogenesis inhibitors is relatively common, with proteinuria occurring in over 60% of patients [2]. Most cases are low-grade, transient, and do not require interventions or dose delays; however, more persistent, severe cases presenting as AKI and nephrotic syndrome can occur [2,3]. Risk factors associated with high-grade proteinuria include increased dose, prolonged administration, pre-existing renal disease, and administration of concurrent chemotherapy [4,5].
The pathophysiology of VEGF inhibitor-induced proteinuria remains unclear. Within a normal kidney, VEGF is produced by podocytes, and VEGF receptors are typically present on the glomerular and peritubular endothelium in addition to mesangial cells [6]. Inhibition of VEGF is thought to cause loss of endothelial fenestrations, podocyte injury and reduce endothelial proliferation, ultimately causing disruption of glomerular membranes [6]. Some cases have also demonstrated subacute thrombotic microangiopathy with endotheliosis and membranoproliferative changes [7]. Another manifestation of nephrotoxicity that is commonly seen is hypertension, occurring in more than a third of patients, which arises due to various mechanisms of renal vascular injury including inhibition of nitric oxide, rarefaction of microvasculature, and neuroendocrine dysregulation [8,9]. It is also hypothesized to increase intraglomerular pressure and ultrafiltration, leading to proteinuria [10].
The lack of clarity surrounding pathophysiology of VEGF inhibitor-induced nephrotoxicity [11] is demonstrated by the heterogeneity of published reports on glomerulopathy and other manifestations, including minimal change disease, collapsing glomerulopathy, membranoproliferative glomerulonephritis, focal segmental glomerulosclerosis, cryoglobulinemic glomerulonephritis, acute tubular necrosis, and interstitial nephritis [7]. Furthermore, worsening kidney disease can further exacerbate hypertension, which may perpetuate AKI [10].
Treatment for low-grade proteinuria usually includes an angiotensin-converting enzyme (ACE) inhibitor or an angiotensin receptor blocker to reduce glomerular filtration pressure, and anti-VEGF treatment can be continued providing stable proteinuria. However, proteinuria may worsen to the nephrotic range (>3 g in 24 h) with nephrotic syndrome. Even after cessation of VEGF inhibitor therapy, there are documented cases of persistent proteinuria [12].
3.1. Angiogenesis Inhibitors Post-Transplant
The role of the VEGF pathway in the pathogenesis of post-transplant complications is poorly understood, with preliminary reports across various organ transplants showing hypothesis-generating results [13,14,15,16]. Upregulation of angiogenesis factors was associated with increased allograft vasculopathy, bronchiolitis obliterans, and recurrence of hepatocellular carcinoma in cardiac, pulmonary, and liver transplants, respectively [13,15,16,17]. In renal transplants, VEGF is thought to be upregulated in acute and chronic rejection, particularly associated with cyclosporine [14].
As the first VEGF inhibitor used for anti-cancer treatment, bevacizumab was approved by the United States Food and Drug Administration in 2004, and is now licensed for use in many cancers [18] (Table 1). In ovarian cancer, patients who were prior recipients of solid organ transplants or who were receiving immunosuppressive therapies were not excluded from randomized bevacizumab trials, but those with pre-existing uncontrolled hypertension or renal dysfunction based on serum creatinine ≥ 1.6 mg/dL or proteinuria > 1 g per 24 h were excluded [19,20,21]. Similarly, in other large randomized angiogenesis inhibitor studies across other tumor sites, prior solid organ transplant or use of immunosuppressants is not an exclusion criterion, aside from studies involving immune checkpoint inhibitors [22]. Identification of patients enrolled in large angiogenesis inhibitor trials who had received prior transplants could potentially make for an interesting post-hoc pooled analysis.
Similarly, reports on angiogenesis inhibition in solid organ transplant patients remain scant in the literature, as highlighted by a review on bevacizumab toxicity by Fenoglio et al [9]. Musri et al. reported a case of colorectal cancer post-renal transplantation with baseline proteinuria, which significantly worsened on administration of intravenous 5-fluorouracil, irinotecan, oxaliplatin, and bevacizumab [23]. Cheungpasitporn et al. described two cases with renal allograft dysfunction following administration of intravitreal bevacizumab, aflibercept, or ranibizumab [24]. Doses of anti-angiogenics were lower but not specified within this report. Although neither case proved causality with anti-VEGF therapy, one was diagnosed with phospholipase A2 receptor-negative membranous nephropathy, and the second revealed acute and chronic antibody-mediated rejection with glomerular thrombi and transplant glomerulopathy. Jonkers and Buren reported a case of worsening IgA nephropathy presenting with nephrotic-range proteinuria post-renal transplantation on sorafenib [25]. These reports highlight the potential severe nephrotoxicity known to be associated with angiogenesis inhibitors; however, there remain few documented positive experiences with angiogenesis inhibitor use in the post-transplant setting. Given the prevalence of nephrotoxicity with these agents, these considerations are particularly prudent for renal transplant recipients, but reports in other organ transplants remain similarly scarce.
3.2. Medication Interactions
Solid organ transplant recipients frequently take maintenance immunosuppressive agents, including but not limited to corticosteroids, calcineurin inhibitors, anti-proliferative agents, and mTOR inhibitors, which are associated with various complications and drug interactions (Table 2).
In the case presented, tacrolimus levels were measured every few months to be within the therapeutic range; this is significant as potential drug interactions between transplant medications and angiogenesis inhibitors involve pharmacokinetic and pharmacodynamic interactions. Pharmacokinetic interactions typically occur due to cytochrome P450 enzyme (CYP) and P-glycoprotein (P-gp) drug transport systems [26,27,28], and risk of competitive metabolism as substrates for the same enzyme or transporter may increase serum levels. Axitinib and sorafenib are CYP3A4 and P-gp substrates, and cabozantinib and pazopanib are substrates and inhibitors of both enzyme systems [29,30,31,32]. Drug interactions are well-described for CYP3A4 substrates cyclosporine, tacrolimus, and sirolimus, with metabolic inhibition leading to increased immunosuppressant concentrations (~20%) [28]. Pharmacodynamic interactions primarily concern cumulative toxicities between these two classes (Table 2) [26,27,33,34]. Interestingly, Onodera et al. reported upon a case of metastatic colorectal cancer post-renal transplant where a patient was administered five cycles of 5-fluorouracil, oxaliplatin, and bevacizumab where although severe proteinuria occurred, serum tacrolimus levels were not affected throughout the course of treatment [35]. This report remains one of the only cases in the literature that demonstrates stability of immunosuppression whilst on bevacizumab post-transplant.
3.3. Long-Term Adverse Events
One of the leading causes of morbidity and mortality in solid organ transplant recipients is malignancy, most commonly non-melanomatous skin cancers [36]. Other malignancies such as colorectal, kidney, and cervical cancers are also prevalent in the post-transplant context, and angiogenesis inhibitors such as bevacizumab are commonly used in metastatic disease [18] (Table 1). Surveillance recommendations within transplant recipients are variable across the globe due to a paucity of robust screening trials [37]. Other long-term complications associated with organ transplantation and prolonged immunosuppressant use include cardiovascular disease, diabetes mellitus, hypertension, and infection associated with cytopenia. In the patient presented, adverse events have not outweighed benefits of ongoing treatment, but this will need close monitoring given the risk of overlapping toxicities as long-term adverse event data remain limited [3,38].
4. Conclusions
Whilst there are minimal data justifying that bevacizumab or other angiogenesis inhibitors are unsafe in the post-transplant setting, there is similarly scarce literature demonstrating safe administration, as in the patient presented. As life expectancy continues to improve with increasing indications for transplantation, long-term risks for malignancy with prolonged immunosuppression are increasingly relevant as a cause of mortality in solid organ transplant recipients. In patients who have undergone renal transplantation, careful consideration of treatment options with risk of nephrotoxicity and close monitoring remains paramount.
Although treating oncologists should remain vigilant about potential drug interactions and overlapping toxicities, these are not necessarily contraindications for agents such as bevacizumab. Treatment decisions should consider the best available evidence, and collating information about toxicity and tolerance from randomized trials and post-approval Phase IV studies would provide detailed information from at-risk subgroups. This calls for a stratified, inclusive approach to allow enrolment of those with chronic diseases and comorbidities in prospective trials, allowing objective assessment of the risk–benefit ratio.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Author Contributions
Conceptualization, L.K., A.M.O.; methodology, L.K., J.D.; software, L.K.; validation, L.K., J.D., J.S., A.K.; resources, L.K.; data curation, L.K., S.L., J.S.; writing—original draft preparation, L.K., J.D.; writing—review and editing, L.K., J.D., K.K., A.K., J.S., S.L., A.M.O.; visualization, L.K., K.K.; supervision, S.L., A.M.O. All authors have read and agreed to the published version of the manuscript.
Funding
This research received no external funding.
Institutional Review Board Statement
As this was a report with fewer than three patients, institutional approval was not sought as per the University Health Network Research Ethics Board guidance document on case reports.
Informed Consent Statement
Informed consent was obtained from all subjects involved in the study and written informed consent has been obtained from the patient to publish this paper.
Data Availability Statement
Not applicable.
Conflicts of Interest
We have read and understood Current Oncology’s policy on disclosing conflicts of interest and declare that we have none.
Abbreviations
VEGFR vascular endothelial growth factor receptor
EGFR epidermal growth factor receptor
FGFR fibroblast growth factor receptor
PDGFR platelet-derived growth factor receptor
SCF stem cell factor
PI3K phosphoinositide 3-kinase
AKT protein kinase B
mTOR mammalian target of rapamycin
RET RET proto-oncogene
AXL AXL receptor tyrosine kinase
TAH–BSO total abdominal hysterectomy and bilateral salpingo-oophorectomy
Ca125 cancer antigen 125
U/ml units per milliliter
AKI acute kidney injury
CYP cytochrome P450
HGSOC high-grade serous ovarian carcinoma
FOLFOX4 5-fluorouracil, leucovorin, and oxaliplatin
IFL irinotecan, leucovorin, and 5-fluorouracil
P-gp P-glycoprotein
Figure 1 Angiogenesis pathways in malignancy and targeted therapies commonly used. Monoclonal antibodies including bevacizumab and ramucirumab and tyrosine kinase inhibitors including sunitinib, sorafenib, lenvatinib, pazopanib, axitinib, cediranib, and cabozantinib have been included (this list is not exhaustive). Receptors including VEGFR, c-kit, epidermal growth factor receptor (EGFR), fibroblast growth factor receptor (FGFR), and PDGFR are displayed here, but certain medications also target other pathways not displayed here such as AXL and RET. The VEGFR pathway intersects with multiple cell signaling pathways, including the PI3K/AKT/mTOR pathway. Created with Biorender.com.
Figure 2 (a) Summary oncology treatment timeline. (b) Ca125 trend from diagnosis to current.
curroncol-28-00064-t001_Table 1Table 1 Food and Drug Administration approved indications for bevacizumab [18].
Cancer Stage Usage
Colorectal Metastatic, first-line 5 mg/kg every two weeks with bolus IFL
10 mg/kg every two weeks with FOLFOX4
Metastatic, recurrent after first-line bevacizumab-containing regimen 5 mg/kg every two weeks, or 7.5 mg/kg every three weeks with fluoropyrimidine–irinotecan, or fluoropyrimidine–oxaliplatin-based chemotherapy
Non-squamous, non-small-cell lung Unresectable, locally advanced, recurrent, or metastatic 15 mg/kg every three weeks with carboplatin and paclitaxel
Glioblastoma Recurrent 10 mg/kg every two weeks
Renal cell Metastatic 10 mg/kg every two weeks with interferon-alfa
Cervical Persistent, recurrent, or metastatic 15 mg/kg every three weeks with paclitaxel and cisplatin, or paclitaxel and topotecan
Epithelial ovarian, fallopian tube, or primary peritoneal III or IV, following surgical resection 15 mg/kg every three weeks with carboplatin and paclitaxel for up to six cycles, followed by 15 mg/kg every three weeks as a single agent for up to 22 cycles
Recurrent, platinum-sensitive 15 mg/kg every three weeks with carboplatin and either paclitaxel (6–8 cycles) or gemcitabine (6–10 cycles) followed by 15 mg/kg every 3 weeks as a single agent
Recurrent, platinum-resistant 10 mg/kg every two weeks with paclitaxel, pegylated liposomal doxorubicin, or topotecan given weekly
15 mg/kg every three weeks with topotecan every three weeks
Hepatocellular Unresectable or metastatic, first-line 15 mg/kg with atezolizumab every three weeks
Abbreviations: mg/kg = milligrams per kilogram; IFL = infusional fluoropyrimidine; FOLFOX4 = 5-fluorouracil, folic acid, and oxaliplatin.
curroncol-28-00064-t002_Table 2Table 2 Drug–drug interactions between post-transplant immunosuppressive medications and angiogenesis inhibitors.
Transplant Medication Potential Interactions with Anti-Angiogenesis Agents [26,27,28,29,30,31,32,33,34]
Cyclosporine
Increased cyclosporine levels and subsequent toxicity due to CYP3A4 and P-gp-mediated drug interactions (e.g., cabozatinib, axitinib, pazopanib, sorafenib, sunitinib)
Tacrolimus
Increased tacrolimus levels and subsequent toxicity due to inhibition of or competition with CYP3A4 metabolism and P-gp-mediated transport (e.g., cabozatinib, axitinib, pazopanib, sorafenib, sunitinib)
Additive impairment of the renal function (e.g., cediranib, axitinib, pazopanib)
Increased risk of QT prolongation with other agents that prolong the QT interval (e.g., cabozantinib, pazopanib, sorafenib, sunitinib)
Mycophenolate mofetil
Exaggerated leukopenia (e.g., ramucirumab, bevacizumab, sunitinib)
Azathioprine
Exaggerated leukopenia (e.g., ramucirumab, bevacizumab, sunitinib)
Sirolimus
Increased tacrolimus levels and subsequent toxicity due to inhibition of or competition with CYP3A4 metabolism and P-gp-mediated transport (e.g., cabozatinib, axitinib, pazopanib, sorafenib, sunitinib)
Additive impairment of the renal function (e.g., cediranib, axitinib, pazopanib)
Additive impairment of wound healing
Everolimus
Increased everolimus levels and subsequent toxicity due to inhibition of CYP3A4 metabolism and P-gp-mediated transport (e.g., cabozantinib, pazopanib)
Additive impairment of the renal function (e.g., cediranib, axitinib, pazopanib)
Additive impairment of wound healing
Corticosteroids
Competitive CYP3A4 metabolism (e.g., prednisone) with other CYP3A4 substrates (e.g., cabozantinib, axitinib, pazopanib, sorafenib, sunitinib)
Abbreviations: CYP3A4 = cytochrome P450 3A4; P-gp = P-glycoprotein. | 5 MG, DAILY | DrugDosageText | CC BY | 33499164 | 18,991,561 | 2021-01-22 |
What was the outcome of reaction 'Acute kidney injury'? | Angiogenesis Inhibitors as Anti-Cancer Therapy Following Renal Transplantation: A Case Report and Review of the Literature.
Solid organ transplant recipients on long-term immunosuppressive medication are at increased risk of developing malignancy, and treatment of advanced cancers with angiogenesis inhibitors in this context has not been widely studied. We present a case of recurrent high-grade serous ovarian carcinoma treated with paclitaxel and bevacizumab in the context of prior renal transplantation where the patient responded well to treatment with controlled toxicities, discussing the potential for increased rates of adverse events and drug interactions in this select population.
1. Introduction
Angiogenesis inhibitors such as vascular endothelial growth factor (VEGF) monoclonal antibodies and tyrosine kinase inhibitors (Figure 1) are standard treatments across various cancer subtypes. In advanced high-grade serous ovarian cancer (HGSOC), bevacizumab is part of the standard of care as maintenance therapy in front-line and recurrent disease [1]. Although the toxicity profile is generally well-managed, nephrotoxicity manifesting as proteinuria remains an important adverse event that requires close monitoring [2]. There is limited literature surrounding angiogenesis inhibitors as anti-cancer treatment in patients who have received solid organ transplants, and thus this report presents a case demonstrating safety of angiogenesis inhibition as anti-cancer therapy in a patient with a stable renal transplant.
2. Case Description
Consent: Fully informed, voluntary, written consent has been obtained to include patient information and publish this report.
Ethics: As this is a case report with fewer than three patients, institutional approval was not sought as per the University Health Network Research Ethics Board guidance document on case reports.
A 47-year-old woman presented with several months of dyspnea and abdominal distension to Princess Margaret Cancer Centre in February 2019 (Figure 2). CT revealed a 13.9 cm pelvic mass with peritoneal carcinomatosis and ascites. Omental biopsy confirmed HGSOC, and she received neoadjuvant platinum-based chemotherapy with excellent tolerance and no renal complications before proceeding to interval debulking in June 2019. There was no visible residual disease, and diagnosis of HGSOC was confirmed, germline and somatic BRCA wild type. She completed three cycles of adjuvant chemotherapy.
Her background was significant for IgA nephropathy, which resulted in progressive chronic kidney disease for 18 years prior to living donor kidney transplant in 2016. Both she and her donor were cytomegalovirus-positive, and she developed cytomegalovirus-associated colitis shortly post-transplantation. Her initial immunosuppression consisted of basiliximab induction followed by tacrolimus, mycophenolic acid, and steroids. She developed antibody-mediated rejection one week post-transplant, which was treated with plasmapheresis, immunoglobulin, and an increase in steroid dose. This was repeated three months later due to biopsy confirming ongoing antibody-mediated rejection. Mycophenolic acid was stopped upon HGSOC diagnosis. She has remained medication-adherent, with regular serum tacrolimus levels within the target range (most recently, 5.3 micrograms/L) and no signs of chronic graft rejection.
Her other comorbidities include diet-controlled, steroid-induced diabetes mellitus; ductal breast carcinoma in situ requiring wide local excision in 2011; asthma; reflux disease; hypertension; and hyperlipidemia. Other medications include prednisone 5 mg daily, acetylsalicylic acid, bisoprolol, trimethoprim–sulfamethoxazole, vitamin D, and inhaled salbutamol as needed.
In December 2019, she developed recurrence in the peritoneum and retroperitoneal lymph nodes, signifying platinum resistance. In January 2020, she commenced weekly paclitaxel 80 mg/m2 with bevacizumab 10 mg/kg every two weeks, and at that time, tacrolimus dose was reduced to 2 mg daily to aim for a serum level of 5 micrograms/L. She continued this therapy for over 6 months, and continued on the same dose of immunosuppression throughout with tacrolimus levels ranging between 3 and 7.6 micrograms/L. Serial imaging and Ca125 confirmed good response to treatment with reduction in size of tumor deposits. Her albumin/creatinine ratio was normal at 0.9 prior to the diagnosis of ovarian cancer, and most recently has been 15.6, signifying microalbuminuria; this has been monitored via urinalysis, which has consistently reported protein as negative or trace. Her most recent estimated glomerular filtration rate was 55 mL/min, similar to pre-diagnosis, and creatinine levels have mostly fluctuated between 95 and 120 umol/L.
Her course has been complicated by grade 2 hypertension (up to 145/95 mmHg) and non-cardiac chest pain, for which amlodipine was switched to ramipril 10 mg daily, and bevacizumab was withheld on two different occasions. Furthermore, she developed a brief period of Kidney Disease: Improving Global Outcomes (KDIGO) stage 1 acute kidney injury (AKI) of pre-renal etiology in September 2020 with creatinine of 160 umol/L, which self-resolved following withdrawal of bevacizumab, and was resumed after a 1-month-long break with no further episodes of kidney injury.
3. Discussion
Nephrotoxicity with angiogenesis inhibitors is relatively common, with proteinuria occurring in over 60% of patients [2]. Most cases are low-grade, transient, and do not require interventions or dose delays; however, more persistent, severe cases presenting as AKI and nephrotic syndrome can occur [2,3]. Risk factors associated with high-grade proteinuria include increased dose, prolonged administration, pre-existing renal disease, and administration of concurrent chemotherapy [4,5].
The pathophysiology of VEGF inhibitor-induced proteinuria remains unclear. Within a normal kidney, VEGF is produced by podocytes, and VEGF receptors are typically present on the glomerular and peritubular endothelium in addition to mesangial cells [6]. Inhibition of VEGF is thought to cause loss of endothelial fenestrations, podocyte injury and reduce endothelial proliferation, ultimately causing disruption of glomerular membranes [6]. Some cases have also demonstrated subacute thrombotic microangiopathy with endotheliosis and membranoproliferative changes [7]. Another manifestation of nephrotoxicity that is commonly seen is hypertension, occurring in more than a third of patients, which arises due to various mechanisms of renal vascular injury including inhibition of nitric oxide, rarefaction of microvasculature, and neuroendocrine dysregulation [8,9]. It is also hypothesized to increase intraglomerular pressure and ultrafiltration, leading to proteinuria [10].
The lack of clarity surrounding pathophysiology of VEGF inhibitor-induced nephrotoxicity [11] is demonstrated by the heterogeneity of published reports on glomerulopathy and other manifestations, including minimal change disease, collapsing glomerulopathy, membranoproliferative glomerulonephritis, focal segmental glomerulosclerosis, cryoglobulinemic glomerulonephritis, acute tubular necrosis, and interstitial nephritis [7]. Furthermore, worsening kidney disease can further exacerbate hypertension, which may perpetuate AKI [10].
Treatment for low-grade proteinuria usually includes an angiotensin-converting enzyme (ACE) inhibitor or an angiotensin receptor blocker to reduce glomerular filtration pressure, and anti-VEGF treatment can be continued providing stable proteinuria. However, proteinuria may worsen to the nephrotic range (>3 g in 24 h) with nephrotic syndrome. Even after cessation of VEGF inhibitor therapy, there are documented cases of persistent proteinuria [12].
3.1. Angiogenesis Inhibitors Post-Transplant
The role of the VEGF pathway in the pathogenesis of post-transplant complications is poorly understood, with preliminary reports across various organ transplants showing hypothesis-generating results [13,14,15,16]. Upregulation of angiogenesis factors was associated with increased allograft vasculopathy, bronchiolitis obliterans, and recurrence of hepatocellular carcinoma in cardiac, pulmonary, and liver transplants, respectively [13,15,16,17]. In renal transplants, VEGF is thought to be upregulated in acute and chronic rejection, particularly associated with cyclosporine [14].
As the first VEGF inhibitor used for anti-cancer treatment, bevacizumab was approved by the United States Food and Drug Administration in 2004, and is now licensed for use in many cancers [18] (Table 1). In ovarian cancer, patients who were prior recipients of solid organ transplants or who were receiving immunosuppressive therapies were not excluded from randomized bevacizumab trials, but those with pre-existing uncontrolled hypertension or renal dysfunction based on serum creatinine ≥ 1.6 mg/dL or proteinuria > 1 g per 24 h were excluded [19,20,21]. Similarly, in other large randomized angiogenesis inhibitor studies across other tumor sites, prior solid organ transplant or use of immunosuppressants is not an exclusion criterion, aside from studies involving immune checkpoint inhibitors [22]. Identification of patients enrolled in large angiogenesis inhibitor trials who had received prior transplants could potentially make for an interesting post-hoc pooled analysis.
Similarly, reports on angiogenesis inhibition in solid organ transplant patients remain scant in the literature, as highlighted by a review on bevacizumab toxicity by Fenoglio et al [9]. Musri et al. reported a case of colorectal cancer post-renal transplantation with baseline proteinuria, which significantly worsened on administration of intravenous 5-fluorouracil, irinotecan, oxaliplatin, and bevacizumab [23]. Cheungpasitporn et al. described two cases with renal allograft dysfunction following administration of intravitreal bevacizumab, aflibercept, or ranibizumab [24]. Doses of anti-angiogenics were lower but not specified within this report. Although neither case proved causality with anti-VEGF therapy, one was diagnosed with phospholipase A2 receptor-negative membranous nephropathy, and the second revealed acute and chronic antibody-mediated rejection with glomerular thrombi and transplant glomerulopathy. Jonkers and Buren reported a case of worsening IgA nephropathy presenting with nephrotic-range proteinuria post-renal transplantation on sorafenib [25]. These reports highlight the potential severe nephrotoxicity known to be associated with angiogenesis inhibitors; however, there remain few documented positive experiences with angiogenesis inhibitor use in the post-transplant setting. Given the prevalence of nephrotoxicity with these agents, these considerations are particularly prudent for renal transplant recipients, but reports in other organ transplants remain similarly scarce.
3.2. Medication Interactions
Solid organ transplant recipients frequently take maintenance immunosuppressive agents, including but not limited to corticosteroids, calcineurin inhibitors, anti-proliferative agents, and mTOR inhibitors, which are associated with various complications and drug interactions (Table 2).
In the case presented, tacrolimus levels were measured every few months to be within the therapeutic range; this is significant as potential drug interactions between transplant medications and angiogenesis inhibitors involve pharmacokinetic and pharmacodynamic interactions. Pharmacokinetic interactions typically occur due to cytochrome P450 enzyme (CYP) and P-glycoprotein (P-gp) drug transport systems [26,27,28], and risk of competitive metabolism as substrates for the same enzyme or transporter may increase serum levels. Axitinib and sorafenib are CYP3A4 and P-gp substrates, and cabozantinib and pazopanib are substrates and inhibitors of both enzyme systems [29,30,31,32]. Drug interactions are well-described for CYP3A4 substrates cyclosporine, tacrolimus, and sirolimus, with metabolic inhibition leading to increased immunosuppressant concentrations (~20%) [28]. Pharmacodynamic interactions primarily concern cumulative toxicities between these two classes (Table 2) [26,27,33,34]. Interestingly, Onodera et al. reported upon a case of metastatic colorectal cancer post-renal transplant where a patient was administered five cycles of 5-fluorouracil, oxaliplatin, and bevacizumab where although severe proteinuria occurred, serum tacrolimus levels were not affected throughout the course of treatment [35]. This report remains one of the only cases in the literature that demonstrates stability of immunosuppression whilst on bevacizumab post-transplant.
3.3. Long-Term Adverse Events
One of the leading causes of morbidity and mortality in solid organ transplant recipients is malignancy, most commonly non-melanomatous skin cancers [36]. Other malignancies such as colorectal, kidney, and cervical cancers are also prevalent in the post-transplant context, and angiogenesis inhibitors such as bevacizumab are commonly used in metastatic disease [18] (Table 1). Surveillance recommendations within transplant recipients are variable across the globe due to a paucity of robust screening trials [37]. Other long-term complications associated with organ transplantation and prolonged immunosuppressant use include cardiovascular disease, diabetes mellitus, hypertension, and infection associated with cytopenia. In the patient presented, adverse events have not outweighed benefits of ongoing treatment, but this will need close monitoring given the risk of overlapping toxicities as long-term adverse event data remain limited [3,38].
4. Conclusions
Whilst there are minimal data justifying that bevacizumab or other angiogenesis inhibitors are unsafe in the post-transplant setting, there is similarly scarce literature demonstrating safe administration, as in the patient presented. As life expectancy continues to improve with increasing indications for transplantation, long-term risks for malignancy with prolonged immunosuppression are increasingly relevant as a cause of mortality in solid organ transplant recipients. In patients who have undergone renal transplantation, careful consideration of treatment options with risk of nephrotoxicity and close monitoring remains paramount.
Although treating oncologists should remain vigilant about potential drug interactions and overlapping toxicities, these are not necessarily contraindications for agents such as bevacizumab. Treatment decisions should consider the best available evidence, and collating information about toxicity and tolerance from randomized trials and post-approval Phase IV studies would provide detailed information from at-risk subgroups. This calls for a stratified, inclusive approach to allow enrolment of those with chronic diseases and comorbidities in prospective trials, allowing objective assessment of the risk–benefit ratio.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Author Contributions
Conceptualization, L.K., A.M.O.; methodology, L.K., J.D.; software, L.K.; validation, L.K., J.D., J.S., A.K.; resources, L.K.; data curation, L.K., S.L., J.S.; writing—original draft preparation, L.K., J.D.; writing—review and editing, L.K., J.D., K.K., A.K., J.S., S.L., A.M.O.; visualization, L.K., K.K.; supervision, S.L., A.M.O. All authors have read and agreed to the published version of the manuscript.
Funding
This research received no external funding.
Institutional Review Board Statement
As this was a report with fewer than three patients, institutional approval was not sought as per the University Health Network Research Ethics Board guidance document on case reports.
Informed Consent Statement
Informed consent was obtained from all subjects involved in the study and written informed consent has been obtained from the patient to publish this paper.
Data Availability Statement
Not applicable.
Conflicts of Interest
We have read and understood Current Oncology’s policy on disclosing conflicts of interest and declare that we have none.
Abbreviations
VEGFR vascular endothelial growth factor receptor
EGFR epidermal growth factor receptor
FGFR fibroblast growth factor receptor
PDGFR platelet-derived growth factor receptor
SCF stem cell factor
PI3K phosphoinositide 3-kinase
AKT protein kinase B
mTOR mammalian target of rapamycin
RET RET proto-oncogene
AXL AXL receptor tyrosine kinase
TAH–BSO total abdominal hysterectomy and bilateral salpingo-oophorectomy
Ca125 cancer antigen 125
U/ml units per milliliter
AKI acute kidney injury
CYP cytochrome P450
HGSOC high-grade serous ovarian carcinoma
FOLFOX4 5-fluorouracil, leucovorin, and oxaliplatin
IFL irinotecan, leucovorin, and 5-fluorouracil
P-gp P-glycoprotein
Figure 1 Angiogenesis pathways in malignancy and targeted therapies commonly used. Monoclonal antibodies including bevacizumab and ramucirumab and tyrosine kinase inhibitors including sunitinib, sorafenib, lenvatinib, pazopanib, axitinib, cediranib, and cabozantinib have been included (this list is not exhaustive). Receptors including VEGFR, c-kit, epidermal growth factor receptor (EGFR), fibroblast growth factor receptor (FGFR), and PDGFR are displayed here, but certain medications also target other pathways not displayed here such as AXL and RET. The VEGFR pathway intersects with multiple cell signaling pathways, including the PI3K/AKT/mTOR pathway. Created with Biorender.com.
Figure 2 (a) Summary oncology treatment timeline. (b) Ca125 trend from diagnosis to current.
curroncol-28-00064-t001_Table 1Table 1 Food and Drug Administration approved indications for bevacizumab [18].
Cancer Stage Usage
Colorectal Metastatic, first-line 5 mg/kg every two weeks with bolus IFL
10 mg/kg every two weeks with FOLFOX4
Metastatic, recurrent after first-line bevacizumab-containing regimen 5 mg/kg every two weeks, or 7.5 mg/kg every three weeks with fluoropyrimidine–irinotecan, or fluoropyrimidine–oxaliplatin-based chemotherapy
Non-squamous, non-small-cell lung Unresectable, locally advanced, recurrent, or metastatic 15 mg/kg every three weeks with carboplatin and paclitaxel
Glioblastoma Recurrent 10 mg/kg every two weeks
Renal cell Metastatic 10 mg/kg every two weeks with interferon-alfa
Cervical Persistent, recurrent, or metastatic 15 mg/kg every three weeks with paclitaxel and cisplatin, or paclitaxel and topotecan
Epithelial ovarian, fallopian tube, or primary peritoneal III or IV, following surgical resection 15 mg/kg every three weeks with carboplatin and paclitaxel for up to six cycles, followed by 15 mg/kg every three weeks as a single agent for up to 22 cycles
Recurrent, platinum-sensitive 15 mg/kg every three weeks with carboplatin and either paclitaxel (6–8 cycles) or gemcitabine (6–10 cycles) followed by 15 mg/kg every 3 weeks as a single agent
Recurrent, platinum-resistant 10 mg/kg every two weeks with paclitaxel, pegylated liposomal doxorubicin, or topotecan given weekly
15 mg/kg every three weeks with topotecan every three weeks
Hepatocellular Unresectable or metastatic, first-line 15 mg/kg with atezolizumab every three weeks
Abbreviations: mg/kg = milligrams per kilogram; IFL = infusional fluoropyrimidine; FOLFOX4 = 5-fluorouracil, folic acid, and oxaliplatin.
curroncol-28-00064-t002_Table 2Table 2 Drug–drug interactions between post-transplant immunosuppressive medications and angiogenesis inhibitors.
Transplant Medication Potential Interactions with Anti-Angiogenesis Agents [26,27,28,29,30,31,32,33,34]
Cyclosporine
Increased cyclosporine levels and subsequent toxicity due to CYP3A4 and P-gp-mediated drug interactions (e.g., cabozatinib, axitinib, pazopanib, sorafenib, sunitinib)
Tacrolimus
Increased tacrolimus levels and subsequent toxicity due to inhibition of or competition with CYP3A4 metabolism and P-gp-mediated transport (e.g., cabozatinib, axitinib, pazopanib, sorafenib, sunitinib)
Additive impairment of the renal function (e.g., cediranib, axitinib, pazopanib)
Increased risk of QT prolongation with other agents that prolong the QT interval (e.g., cabozantinib, pazopanib, sorafenib, sunitinib)
Mycophenolate mofetil
Exaggerated leukopenia (e.g., ramucirumab, bevacizumab, sunitinib)
Azathioprine
Exaggerated leukopenia (e.g., ramucirumab, bevacizumab, sunitinib)
Sirolimus
Increased tacrolimus levels and subsequent toxicity due to inhibition of or competition with CYP3A4 metabolism and P-gp-mediated transport (e.g., cabozatinib, axitinib, pazopanib, sorafenib, sunitinib)
Additive impairment of the renal function (e.g., cediranib, axitinib, pazopanib)
Additive impairment of wound healing
Everolimus
Increased everolimus levels and subsequent toxicity due to inhibition of CYP3A4 metabolism and P-gp-mediated transport (e.g., cabozantinib, pazopanib)
Additive impairment of the renal function (e.g., cediranib, axitinib, pazopanib)
Additive impairment of wound healing
Corticosteroids
Competitive CYP3A4 metabolism (e.g., prednisone) with other CYP3A4 substrates (e.g., cabozantinib, axitinib, pazopanib, sorafenib, sunitinib)
Abbreviations: CYP3A4 = cytochrome P450 3A4; P-gp = P-glycoprotein. | Recovered | ReactionOutcome | CC BY | 33499164 | 19,169,975 | 2021-01-22 |
What was the outcome of reaction 'Ovarian cancer'? | Angiogenesis Inhibitors as Anti-Cancer Therapy Following Renal Transplantation: A Case Report and Review of the Literature.
Solid organ transplant recipients on long-term immunosuppressive medication are at increased risk of developing malignancy, and treatment of advanced cancers with angiogenesis inhibitors in this context has not been widely studied. We present a case of recurrent high-grade serous ovarian carcinoma treated with paclitaxel and bevacizumab in the context of prior renal transplantation where the patient responded well to treatment with controlled toxicities, discussing the potential for increased rates of adverse events and drug interactions in this select population.
1. Introduction
Angiogenesis inhibitors such as vascular endothelial growth factor (VEGF) monoclonal antibodies and tyrosine kinase inhibitors (Figure 1) are standard treatments across various cancer subtypes. In advanced high-grade serous ovarian cancer (HGSOC), bevacizumab is part of the standard of care as maintenance therapy in front-line and recurrent disease [1]. Although the toxicity profile is generally well-managed, nephrotoxicity manifesting as proteinuria remains an important adverse event that requires close monitoring [2]. There is limited literature surrounding angiogenesis inhibitors as anti-cancer treatment in patients who have received solid organ transplants, and thus this report presents a case demonstrating safety of angiogenesis inhibition as anti-cancer therapy in a patient with a stable renal transplant.
2. Case Description
Consent: Fully informed, voluntary, written consent has been obtained to include patient information and publish this report.
Ethics: As this is a case report with fewer than three patients, institutional approval was not sought as per the University Health Network Research Ethics Board guidance document on case reports.
A 47-year-old woman presented with several months of dyspnea and abdominal distension to Princess Margaret Cancer Centre in February 2019 (Figure 2). CT revealed a 13.9 cm pelvic mass with peritoneal carcinomatosis and ascites. Omental biopsy confirmed HGSOC, and she received neoadjuvant platinum-based chemotherapy with excellent tolerance and no renal complications before proceeding to interval debulking in June 2019. There was no visible residual disease, and diagnosis of HGSOC was confirmed, germline and somatic BRCA wild type. She completed three cycles of adjuvant chemotherapy.
Her background was significant for IgA nephropathy, which resulted in progressive chronic kidney disease for 18 years prior to living donor kidney transplant in 2016. Both she and her donor were cytomegalovirus-positive, and she developed cytomegalovirus-associated colitis shortly post-transplantation. Her initial immunosuppression consisted of basiliximab induction followed by tacrolimus, mycophenolic acid, and steroids. She developed antibody-mediated rejection one week post-transplant, which was treated with plasmapheresis, immunoglobulin, and an increase in steroid dose. This was repeated three months later due to biopsy confirming ongoing antibody-mediated rejection. Mycophenolic acid was stopped upon HGSOC diagnosis. She has remained medication-adherent, with regular serum tacrolimus levels within the target range (most recently, 5.3 micrograms/L) and no signs of chronic graft rejection.
Her other comorbidities include diet-controlled, steroid-induced diabetes mellitus; ductal breast carcinoma in situ requiring wide local excision in 2011; asthma; reflux disease; hypertension; and hyperlipidemia. Other medications include prednisone 5 mg daily, acetylsalicylic acid, bisoprolol, trimethoprim–sulfamethoxazole, vitamin D, and inhaled salbutamol as needed.
In December 2019, she developed recurrence in the peritoneum and retroperitoneal lymph nodes, signifying platinum resistance. In January 2020, she commenced weekly paclitaxel 80 mg/m2 with bevacizumab 10 mg/kg every two weeks, and at that time, tacrolimus dose was reduced to 2 mg daily to aim for a serum level of 5 micrograms/L. She continued this therapy for over 6 months, and continued on the same dose of immunosuppression throughout with tacrolimus levels ranging between 3 and 7.6 micrograms/L. Serial imaging and Ca125 confirmed good response to treatment with reduction in size of tumor deposits. Her albumin/creatinine ratio was normal at 0.9 prior to the diagnosis of ovarian cancer, and most recently has been 15.6, signifying microalbuminuria; this has been monitored via urinalysis, which has consistently reported protein as negative or trace. Her most recent estimated glomerular filtration rate was 55 mL/min, similar to pre-diagnosis, and creatinine levels have mostly fluctuated between 95 and 120 umol/L.
Her course has been complicated by grade 2 hypertension (up to 145/95 mmHg) and non-cardiac chest pain, for which amlodipine was switched to ramipril 10 mg daily, and bevacizumab was withheld on two different occasions. Furthermore, she developed a brief period of Kidney Disease: Improving Global Outcomes (KDIGO) stage 1 acute kidney injury (AKI) of pre-renal etiology in September 2020 with creatinine of 160 umol/L, which self-resolved following withdrawal of bevacizumab, and was resumed after a 1-month-long break with no further episodes of kidney injury.
3. Discussion
Nephrotoxicity with angiogenesis inhibitors is relatively common, with proteinuria occurring in over 60% of patients [2]. Most cases are low-grade, transient, and do not require interventions or dose delays; however, more persistent, severe cases presenting as AKI and nephrotic syndrome can occur [2,3]. Risk factors associated with high-grade proteinuria include increased dose, prolonged administration, pre-existing renal disease, and administration of concurrent chemotherapy [4,5].
The pathophysiology of VEGF inhibitor-induced proteinuria remains unclear. Within a normal kidney, VEGF is produced by podocytes, and VEGF receptors are typically present on the glomerular and peritubular endothelium in addition to mesangial cells [6]. Inhibition of VEGF is thought to cause loss of endothelial fenestrations, podocyte injury and reduce endothelial proliferation, ultimately causing disruption of glomerular membranes [6]. Some cases have also demonstrated subacute thrombotic microangiopathy with endotheliosis and membranoproliferative changes [7]. Another manifestation of nephrotoxicity that is commonly seen is hypertension, occurring in more than a third of patients, which arises due to various mechanisms of renal vascular injury including inhibition of nitric oxide, rarefaction of microvasculature, and neuroendocrine dysregulation [8,9]. It is also hypothesized to increase intraglomerular pressure and ultrafiltration, leading to proteinuria [10].
The lack of clarity surrounding pathophysiology of VEGF inhibitor-induced nephrotoxicity [11] is demonstrated by the heterogeneity of published reports on glomerulopathy and other manifestations, including minimal change disease, collapsing glomerulopathy, membranoproliferative glomerulonephritis, focal segmental glomerulosclerosis, cryoglobulinemic glomerulonephritis, acute tubular necrosis, and interstitial nephritis [7]. Furthermore, worsening kidney disease can further exacerbate hypertension, which may perpetuate AKI [10].
Treatment for low-grade proteinuria usually includes an angiotensin-converting enzyme (ACE) inhibitor or an angiotensin receptor blocker to reduce glomerular filtration pressure, and anti-VEGF treatment can be continued providing stable proteinuria. However, proteinuria may worsen to the nephrotic range (>3 g in 24 h) with nephrotic syndrome. Even after cessation of VEGF inhibitor therapy, there are documented cases of persistent proteinuria [12].
3.1. Angiogenesis Inhibitors Post-Transplant
The role of the VEGF pathway in the pathogenesis of post-transplant complications is poorly understood, with preliminary reports across various organ transplants showing hypothesis-generating results [13,14,15,16]. Upregulation of angiogenesis factors was associated with increased allograft vasculopathy, bronchiolitis obliterans, and recurrence of hepatocellular carcinoma in cardiac, pulmonary, and liver transplants, respectively [13,15,16,17]. In renal transplants, VEGF is thought to be upregulated in acute and chronic rejection, particularly associated with cyclosporine [14].
As the first VEGF inhibitor used for anti-cancer treatment, bevacizumab was approved by the United States Food and Drug Administration in 2004, and is now licensed for use in many cancers [18] (Table 1). In ovarian cancer, patients who were prior recipients of solid organ transplants or who were receiving immunosuppressive therapies were not excluded from randomized bevacizumab trials, but those with pre-existing uncontrolled hypertension or renal dysfunction based on serum creatinine ≥ 1.6 mg/dL or proteinuria > 1 g per 24 h were excluded [19,20,21]. Similarly, in other large randomized angiogenesis inhibitor studies across other tumor sites, prior solid organ transplant or use of immunosuppressants is not an exclusion criterion, aside from studies involving immune checkpoint inhibitors [22]. Identification of patients enrolled in large angiogenesis inhibitor trials who had received prior transplants could potentially make for an interesting post-hoc pooled analysis.
Similarly, reports on angiogenesis inhibition in solid organ transplant patients remain scant in the literature, as highlighted by a review on bevacizumab toxicity by Fenoglio et al [9]. Musri et al. reported a case of colorectal cancer post-renal transplantation with baseline proteinuria, which significantly worsened on administration of intravenous 5-fluorouracil, irinotecan, oxaliplatin, and bevacizumab [23]. Cheungpasitporn et al. described two cases with renal allograft dysfunction following administration of intravitreal bevacizumab, aflibercept, or ranibizumab [24]. Doses of anti-angiogenics were lower but not specified within this report. Although neither case proved causality with anti-VEGF therapy, one was diagnosed with phospholipase A2 receptor-negative membranous nephropathy, and the second revealed acute and chronic antibody-mediated rejection with glomerular thrombi and transplant glomerulopathy. Jonkers and Buren reported a case of worsening IgA nephropathy presenting with nephrotic-range proteinuria post-renal transplantation on sorafenib [25]. These reports highlight the potential severe nephrotoxicity known to be associated with angiogenesis inhibitors; however, there remain few documented positive experiences with angiogenesis inhibitor use in the post-transplant setting. Given the prevalence of nephrotoxicity with these agents, these considerations are particularly prudent for renal transplant recipients, but reports in other organ transplants remain similarly scarce.
3.2. Medication Interactions
Solid organ transplant recipients frequently take maintenance immunosuppressive agents, including but not limited to corticosteroids, calcineurin inhibitors, anti-proliferative agents, and mTOR inhibitors, which are associated with various complications and drug interactions (Table 2).
In the case presented, tacrolimus levels were measured every few months to be within the therapeutic range; this is significant as potential drug interactions between transplant medications and angiogenesis inhibitors involve pharmacokinetic and pharmacodynamic interactions. Pharmacokinetic interactions typically occur due to cytochrome P450 enzyme (CYP) and P-glycoprotein (P-gp) drug transport systems [26,27,28], and risk of competitive metabolism as substrates for the same enzyme or transporter may increase serum levels. Axitinib and sorafenib are CYP3A4 and P-gp substrates, and cabozantinib and pazopanib are substrates and inhibitors of both enzyme systems [29,30,31,32]. Drug interactions are well-described for CYP3A4 substrates cyclosporine, tacrolimus, and sirolimus, with metabolic inhibition leading to increased immunosuppressant concentrations (~20%) [28]. Pharmacodynamic interactions primarily concern cumulative toxicities between these two classes (Table 2) [26,27,33,34]. Interestingly, Onodera et al. reported upon a case of metastatic colorectal cancer post-renal transplant where a patient was administered five cycles of 5-fluorouracil, oxaliplatin, and bevacizumab where although severe proteinuria occurred, serum tacrolimus levels were not affected throughout the course of treatment [35]. This report remains one of the only cases in the literature that demonstrates stability of immunosuppression whilst on bevacizumab post-transplant.
3.3. Long-Term Adverse Events
One of the leading causes of morbidity and mortality in solid organ transplant recipients is malignancy, most commonly non-melanomatous skin cancers [36]. Other malignancies such as colorectal, kidney, and cervical cancers are also prevalent in the post-transplant context, and angiogenesis inhibitors such as bevacizumab are commonly used in metastatic disease [18] (Table 1). Surveillance recommendations within transplant recipients are variable across the globe due to a paucity of robust screening trials [37]. Other long-term complications associated with organ transplantation and prolonged immunosuppressant use include cardiovascular disease, diabetes mellitus, hypertension, and infection associated with cytopenia. In the patient presented, adverse events have not outweighed benefits of ongoing treatment, but this will need close monitoring given the risk of overlapping toxicities as long-term adverse event data remain limited [3,38].
4. Conclusions
Whilst there are minimal data justifying that bevacizumab or other angiogenesis inhibitors are unsafe in the post-transplant setting, there is similarly scarce literature demonstrating safe administration, as in the patient presented. As life expectancy continues to improve with increasing indications for transplantation, long-term risks for malignancy with prolonged immunosuppression are increasingly relevant as a cause of mortality in solid organ transplant recipients. In patients who have undergone renal transplantation, careful consideration of treatment options with risk of nephrotoxicity and close monitoring remains paramount.
Although treating oncologists should remain vigilant about potential drug interactions and overlapping toxicities, these are not necessarily contraindications for agents such as bevacizumab. Treatment decisions should consider the best available evidence, and collating information about toxicity and tolerance from randomized trials and post-approval Phase IV studies would provide detailed information from at-risk subgroups. This calls for a stratified, inclusive approach to allow enrolment of those with chronic diseases and comorbidities in prospective trials, allowing objective assessment of the risk–benefit ratio.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Author Contributions
Conceptualization, L.K., A.M.O.; methodology, L.K., J.D.; software, L.K.; validation, L.K., J.D., J.S., A.K.; resources, L.K.; data curation, L.K., S.L., J.S.; writing—original draft preparation, L.K., J.D.; writing—review and editing, L.K., J.D., K.K., A.K., J.S., S.L., A.M.O.; visualization, L.K., K.K.; supervision, S.L., A.M.O. All authors have read and agreed to the published version of the manuscript.
Funding
This research received no external funding.
Institutional Review Board Statement
As this was a report with fewer than three patients, institutional approval was not sought as per the University Health Network Research Ethics Board guidance document on case reports.
Informed Consent Statement
Informed consent was obtained from all subjects involved in the study and written informed consent has been obtained from the patient to publish this paper.
Data Availability Statement
Not applicable.
Conflicts of Interest
We have read and understood Current Oncology’s policy on disclosing conflicts of interest and declare that we have none.
Abbreviations
VEGFR vascular endothelial growth factor receptor
EGFR epidermal growth factor receptor
FGFR fibroblast growth factor receptor
PDGFR platelet-derived growth factor receptor
SCF stem cell factor
PI3K phosphoinositide 3-kinase
AKT protein kinase B
mTOR mammalian target of rapamycin
RET RET proto-oncogene
AXL AXL receptor tyrosine kinase
TAH–BSO total abdominal hysterectomy and bilateral salpingo-oophorectomy
Ca125 cancer antigen 125
U/ml units per milliliter
AKI acute kidney injury
CYP cytochrome P450
HGSOC high-grade serous ovarian carcinoma
FOLFOX4 5-fluorouracil, leucovorin, and oxaliplatin
IFL irinotecan, leucovorin, and 5-fluorouracil
P-gp P-glycoprotein
Figure 1 Angiogenesis pathways in malignancy and targeted therapies commonly used. Monoclonal antibodies including bevacizumab and ramucirumab and tyrosine kinase inhibitors including sunitinib, sorafenib, lenvatinib, pazopanib, axitinib, cediranib, and cabozantinib have been included (this list is not exhaustive). Receptors including VEGFR, c-kit, epidermal growth factor receptor (EGFR), fibroblast growth factor receptor (FGFR), and PDGFR are displayed here, but certain medications also target other pathways not displayed here such as AXL and RET. The VEGFR pathway intersects with multiple cell signaling pathways, including the PI3K/AKT/mTOR pathway. Created with Biorender.com.
Figure 2 (a) Summary oncology treatment timeline. (b) Ca125 trend from diagnosis to current.
curroncol-28-00064-t001_Table 1Table 1 Food and Drug Administration approved indications for bevacizumab [18].
Cancer Stage Usage
Colorectal Metastatic, first-line 5 mg/kg every two weeks with bolus IFL
10 mg/kg every two weeks with FOLFOX4
Metastatic, recurrent after first-line bevacizumab-containing regimen 5 mg/kg every two weeks, or 7.5 mg/kg every three weeks with fluoropyrimidine–irinotecan, or fluoropyrimidine–oxaliplatin-based chemotherapy
Non-squamous, non-small-cell lung Unresectable, locally advanced, recurrent, or metastatic 15 mg/kg every three weeks with carboplatin and paclitaxel
Glioblastoma Recurrent 10 mg/kg every two weeks
Renal cell Metastatic 10 mg/kg every two weeks with interferon-alfa
Cervical Persistent, recurrent, or metastatic 15 mg/kg every three weeks with paclitaxel and cisplatin, or paclitaxel and topotecan
Epithelial ovarian, fallopian tube, or primary peritoneal III or IV, following surgical resection 15 mg/kg every three weeks with carboplatin and paclitaxel for up to six cycles, followed by 15 mg/kg every three weeks as a single agent for up to 22 cycles
Recurrent, platinum-sensitive 15 mg/kg every three weeks with carboplatin and either paclitaxel (6–8 cycles) or gemcitabine (6–10 cycles) followed by 15 mg/kg every 3 weeks as a single agent
Recurrent, platinum-resistant 10 mg/kg every two weeks with paclitaxel, pegylated liposomal doxorubicin, or topotecan given weekly
15 mg/kg every three weeks with topotecan every three weeks
Hepatocellular Unresectable or metastatic, first-line 15 mg/kg with atezolizumab every three weeks
Abbreviations: mg/kg = milligrams per kilogram; IFL = infusional fluoropyrimidine; FOLFOX4 = 5-fluorouracil, folic acid, and oxaliplatin.
curroncol-28-00064-t002_Table 2Table 2 Drug–drug interactions between post-transplant immunosuppressive medications and angiogenesis inhibitors.
Transplant Medication Potential Interactions with Anti-Angiogenesis Agents [26,27,28,29,30,31,32,33,34]
Cyclosporine
Increased cyclosporine levels and subsequent toxicity due to CYP3A4 and P-gp-mediated drug interactions (e.g., cabozatinib, axitinib, pazopanib, sorafenib, sunitinib)
Tacrolimus
Increased tacrolimus levels and subsequent toxicity due to inhibition of or competition with CYP3A4 metabolism and P-gp-mediated transport (e.g., cabozatinib, axitinib, pazopanib, sorafenib, sunitinib)
Additive impairment of the renal function (e.g., cediranib, axitinib, pazopanib)
Increased risk of QT prolongation with other agents that prolong the QT interval (e.g., cabozantinib, pazopanib, sorafenib, sunitinib)
Mycophenolate mofetil
Exaggerated leukopenia (e.g., ramucirumab, bevacizumab, sunitinib)
Azathioprine
Exaggerated leukopenia (e.g., ramucirumab, bevacizumab, sunitinib)
Sirolimus
Increased tacrolimus levels and subsequent toxicity due to inhibition of or competition with CYP3A4 metabolism and P-gp-mediated transport (e.g., cabozatinib, axitinib, pazopanib, sorafenib, sunitinib)
Additive impairment of the renal function (e.g., cediranib, axitinib, pazopanib)
Additive impairment of wound healing
Everolimus
Increased everolimus levels and subsequent toxicity due to inhibition of CYP3A4 metabolism and P-gp-mediated transport (e.g., cabozantinib, pazopanib)
Additive impairment of the renal function (e.g., cediranib, axitinib, pazopanib)
Additive impairment of wound healing
Corticosteroids
Competitive CYP3A4 metabolism (e.g., prednisone) with other CYP3A4 substrates (e.g., cabozantinib, axitinib, pazopanib, sorafenib, sunitinib)
Abbreviations: CYP3A4 = cytochrome P450 3A4; P-gp = P-glycoprotein. | Recovering | ReactionOutcome | CC BY | 33499164 | 19,169,975 | 2021-01-22 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Bradycardia'. | Cardiac arrest caused by accidental severe hypothermia and myocardial infarction during general anesthesia.
Therapeutic hypothermia is often used for traumatic brain injury because of its neuroprotective effect and decreased secondary brain injury. However, this procedure lacks clinical evidence supporting its efficacy, and adverse outcomes have been reported during general anesthesia. A 61-year-old man with a history of percutaneous coronary intervention (PCI) was admitted with traumatic brain injury. Immediately after admission, he underwent mild therapeutic hypothermia with a target temperature of 33.0°C for neuroprotection. During general anesthesia for emergency surgery because he developed a mass effect, hypothermic cardiac arrest occurred following an additional decrease in the core body temperature. Moreover, myocardial infarction caused by restenosis of the previous PCI lesion also contributed to the cardiac arrest. Although the patient recovered spontaneous circulation after an hour-long cardiopulmonary resuscitation with rewarming, he eventually died of subsequent repetitive cardiac arrests. When anesthetizing patients undergoing therapeutic hypothermia, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.
Introduction
Therapeutic hypothermia is an attractive strategy for traumatic brain injury (TBI), which demonstrates neuroprotective effects and decreases secondary brain injury.1 Despite the lack of clinical evidence regarding efficacy, therapeutic hypothermia after TBI has been used for at least half a century because of its potential benefits, which are supported by numerous preclinical studies.2,3 However, there have been no reports on the danger of unintentional severe hypothermia during general anesthesia caused by the additional drop in temperature of patients undergoing therapeutic hypothermia. When anesthetizing patients with comorbidities, such as ischemic heart disease, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.4,5
Case report
A 61-year-old man (height, 168 cm; weight, 70 kg) presented to our emergency room (ER) with head trauma after falling on a cement floor from a 5-meter-tall construction site. Upon arrival at the ER, the patient had a Glasgow coma scale score of 11 and the following vital signs: blood pressure (BP), 147/83 mmHg; heart rate (HR), 86 beats per minute (bpm); respiratory rate, 24 breaths per minute; body temperature from the auditory canal, 36.8°C; and percutaneous oxygen saturation (SpO2), 97%. Upon discovering signs of acute subdural hemorrhage, subarachnoid hemorrhage, and intraventricular hemorrhage from a brain computed tomography (CT) scan performed in the ER, we admitted the patient to the intensive care unit (ICU) for close monitoring. After approximately 6 hours, we performed follow-up brain CT because the patient’s mental status had deteriorated from drowsy to stupor. Upon observing a mass effect characterized by an increase in the hemorrhage volume and a left midline shift on the CT images, the patient was scheduled for immediate intubation and emergency decompressive craniectomy (Figure 1).
Figure 1. Brain computed tomography (CT) images taken 6 hours after intensive care unit admission. (a) Axial view showing acute subarachnoid hemorrhage in the peri-mesencephalic cistern (arrow A) and subdural hemorrhage in the right inferior temporal (arrow B) and right parietal (arrow C) regions. (b) Axial view showing acute subdural hemorrhage in the right temporal region (black arrow) and left midline shift.
Although the patient had a history of ischemic heart disease with percutaneous coronary intervention (PCI), it was difficult to obtain accurate information because there was no relevant data in the patient’s medical records from the authors’ hospital, and his drowsy state interfered with detailed interviews. According to the chest CT scan performed immediately after admission, coronary stents had been inserted at the bifurcation site of the left main coronary artery (Figure 2). It also appeared that the inserted stents were drug-eluting stents (DES) considering that he had received dual antiplatelet therapy (DAPT) with aspirin and clopidogrel (Plavix®; Sanofi, Bridgewater, NJ, USA) a switching platelet P2Y12 inhibitor, for approximately 4 months. However, DAPT was discontinued owing to the worsening cerebral hemorrhage after hospitalization; bridging antiplatelet therapy was also not performed. Before the surgery, the patient showed a normal electrocardiogram (ECG), and cardiac biomarker testing revealed a troponin-T concentration of 0.006 µg/L, which was within normal limits (Figure 3). The patient demonstrated satisfactory functional capacity to be able to work at a construction site just before the trauma. Therefore, an additional preoperative cardiac evaluation was not considered necessary.
Figure 2. Chest computed tomography (CT) images showing coronary stents in the main bifurcation of the left coronary arteries (white arrow in (a) and (b)).
Figure 3. Preoperative electrocardiogram of the patient showing normal sinus rhythm with a heart rate of 95 bpm.
bpm, beats per minute.
Immediately after admission to the ICU, the patient’s bladder temperature was 38.0°C. Using a hydrogel-coated water-circulating pad (Arctic Sun®; Medivance Inc., Louisville, CO, USA), which is a surface cooling device, the patient underwent mild therapeutic hypothermia with a target temperature of 33.0°C. During therapeutic hypothermia, sedation and shivering control were performed using continuous rate infusions of midazolam at 5 mg/hour and remifentanil at 0.2 µg/kg/minute. The patient’s bladder temperature was 33.4°C just before he was transferred to the operating room. Preoperative laboratory tests showed no unusual findings except for mildly elevated aspartate aminotransferase (73 U/L) and alanine aminotransferase (51 U/L) concentrations.
The patient’s vital signs upon arrival at the operating room were a BP of 130/70 mmHg and SpO2 of 100%. Lead II of the ECG was monitored, and there were no unusual findings except sinus tachycardia with an HR of 100 bpm on the initial ECG. Arterial BP (ABP), central venous pressure (CVP), and esophageal temperature were monitored from the right radial artery, right subclavian vein, and lower third of the esophagus using a temperature sensor attached to the esophageal stethoscope, respectively. The initial ABP, CVP, and esophageal temperature were 140/85 mmHg, 6 mmHg, and 33.0°C, respectively. Arterial blood gas analysis was performed in the arterial line, with values calculated as for a body temperature of 37°C, which revealed the following: pH, 7.400; partial pressure of carbon dioxide (PCO2), 40 mmHg; partial pressure of oxygen (PO2), 174 mmHg; bicarbonate (HCO3), 24.8 mmol/L; base excess, 0.0; lactate, 0.17 mmol/L; ionized calcium, 1.04 mmol/L; and potassium, 3.3 mmol/L. Total intravenous (IV) anesthesia was performed using a target-controlled infusion of propofol and remifentanil in accordance with the Schnider6 and Minto7 models, respectively. Propofol and remifentanil were infused continuously with effect-site concentrations (Ce) of 1.5 µg/mL and 1.5 ng/mL, respectively. Concurrently, 50 mg of rocuronium was administered by IV bolus injection for neuromuscular blockade. Mechanical ventilation was adjusted to maintain the end-tidal carbon dioxide value at 30 to 33 mmHg.
As ABP decreased to 78/42 mmHg and CVP decreased to 4 mmHg within 10 minutes after anesthesia, the Ce of propofol and remifentanil were lowered to 0.8 µg/mL and 1.0 ng/mL, respectively, and approximately 1 L of balanced crystalloid fluid loading was performed. However, the patient’s BP did not recover, and a dopamine continuous infusion (5 µg/kg/minute) was initiated. When a surgical incision was made 10 minutes after beginning the dopamine infusion, the patient’s ABP and HR increased to 115/70 mmHg and 120 bpm, respectively, and marked ST-segment depression (−2.4 mm) was found, which was not observed on the preoperative or initial ECGs (Figure 4). Subsequently, the depth of anesthesia was increased by increasing the Ce of propofol to 1.0 to 1.2 µg/mL, followed by a continuous infusion of isosorbide dinitrate (1 µg/kg/minute). While HR was maintained at 110 to 120 bpm, ABP gradually decreased to 70/40 mmHg after approximately 5 minutes. To treat the hypotension, isosorbide dinitrate was discontinued, and a bolus injection of 100 µg phenylephrine was administered, after which, ABP increased temporarily to 95/60 mmHg. Subsequently, HR decreased suddenly and rapidly, and ventricular fibrillation occurred. The anesthesiologists immediately discontinued the surgery and performed cardiopulmonary resuscitation (CPR). The patient’s esophageal temperature at the time was 32.0°C.
Figure 4. Intraoperative electrocardiogram of the patient showing sinus tachycardia with significant ST depression. The ST depression on the monitor represented approximately −2.4 at the time.
Chest compressions were performed with periodic defibrillation with 200 J, and epinephrine injections were administered every 3 minutes; these measures elicited no response. During CPR, ECG monitoring showed temporary severe bradycardia (HR, 10–30 bpm). However, this was considered pulseless electrical activity because no pulse was detected in the patient’s femoral artery. Soon afterward, there was a rapid transition to ventricular fibrillation, and CPR was repeated. At the time, the patient’s esophageal temperature dropped to 30.0°C. According to the refractory state after the defibrillation and drug injections, we considered that the cardiac arrest was caused by hypothermia. To reverse the cause, a forced-air warming device and a warming blanket (Bair Hugger®; Arizant Healthcare Inc., Eden Prairie, MN, USA) were used for surface warming, the IV fluids were warmed, and surgical wound closure and dressing were performed rapidly to prevent surface heat loss. The first arterial blood gas analysis after ventricular fibrillation, calculated as for a body temperature of 37°C, showed the following: pH, 7.690; PCO2, 15 mmHg; PO2, 260 mmHg; HCO3, 18.1 mmol/L; base excess, −1.2; lactate, 0.33 mmol/L; ionized calcium, 0.88 mmol/L; and potassium, 5.0 mmol/L. Follow-up arterial blood gas analyses were repeated periodically. As we continued to perform CPR, the patient developed severe metabolic acidosis and hypocalcemia characterized by a pH of 6.840, HCO3 of 15.4 mmol/L, and ionized calcium of 0.66 mmol/L. To correct the abnormal findings, we administered appropriate amounts of sodium bicarbonate and calcium chloride. After an hour-long CPR and rewarming, the patient’s esophageal temperature recovered to 33.0°C, which was followed by observing sinus rhythm characterized by an HR of 75 bpm on an ECG monitor and detecting a pulse in the femoral artery with a characteristic ABP waveform that eventually led to the recovery of BP to 130/55 mmHg. Upon confirming recovery of spontaneous circulation (ROSC), the patient was transported to the ICU while we administered dopamine (10 µg/kg/minute), norepinephrine (0.1 µg/kg/minute), and vasopressin (2 U/hour). ABP, HR, and esophageal temperature during anesthesia are shown in Figure 5.
Figure 5. Changes in vital signs (blood pressure (BP), heart rate (HR), and esophageal temperature) over time (5-minute intervals).
bpm, beats per minute.
Cardiac biomarker testing performed after the patient’s transfer to the ICU showed an increased troponin-T level of 1.432 µg/L. Portable Doppler transthoracic echocardiography (TTE) revealed total occlusion of the left main coronary artery, which had undergone PCI previously. Because emergency coronary artery bypass grafting (CABG) was impossible at the time, repeated PCI was considered an alternative treatment option. However, even after his transfer to the ICU, the patient experienced repetitive ventricular fibrillation. Four hours after the patient’s transfer to the ICU, the patient’s guardian requested that CPR should be stopped. Without resuscitation, the patient died.
Discussion
In this case, the patient, who was receiving therapeutic hypothermia, underwent emergency craniectomy owing to worsening brain hemorrhage, and cardiac arrest occurred during general anesthesia. We deduced that the cardiac arrest was owing to the additional decrease in the patient’s core body temperature and myocardial infarction caused by restenosis of the previous PCI-treated lesion.
Therapeutic hypothermia is used to improve neurological outcomes in cardiac arrest, TBI, stroke, and hypoxic–ischemic encephalopathy.1,4 Studies on the efficacy of therapeutic hypothermia for TBI have been less convincing, and they show conflicting results. However, hypothermia is consistently used as an option for TBI because it reduces intracranial pressure and the cerebral metabolic rate, thereby protecting neuronal function in the brain and decreasing the wide range of injuries or reparative processes by attenuating temperature-sensitive mechanisms, such as excitotoxicity, free radical generation, apoptosis, and inflammation.1,2
The cardiovascular effects associated with hypothermia are complex. Hypothermia affects the electrical conduction of myocardial cells, which can induce hypotension or bradycardia and even increase the risk of ventricular fibrillation by prolonging repolarization.4,8 However, mild hypothermia of 32.0°C to 35.0°C demonstrates cardioprotective effects by decreasing myocardial oxygen demand and increasing myocardial perfusion through vasodilation of coronary vessels.9 Therefore, there has been a growing interest in the clinical use of mild therapeutic hypothermia as an adjunctive therapy during ischemic heart disease.10 Our patient received mild therapeutic hypothermia with a target core temperature of 33.0°C, which may have yielded cardioprotective as well as neuroprotective benefits.11
During general anesthesia, hypothermia is induced following an increase in heat loss caused by peripheral vasodilation after the injection of anesthetic drugs, heat redistribution from the core to the periphery, and impaired normal thermoregulatory responses, such as vasoconstriction or shivering.5 In our patient, the cold operating room temperature and irrigating an uncovered surgical area in the brain with cold saline also may have contributed to the additional core body temperature drop in addition to the effects of general anesthesia.12 However, in our case, neurosurgeons requested that anesthesiologists not perform warming during anesthesia. Instead, esophageal temperature and signs that could result from hypothermia, such as ECG changes, were carefully monitored. The esophageal temperature decreased from 33.0°C, which was measured immediately after entering the operation room, to 32.0°C at the time of cardiac arrest. The esophageal temperature further decreased to a minimum of 30.0°C during CPR. With rewarming, the temperature recovered to 33.0°C and finally, ROSC was achieved. Considering that chest compressions, drugs, and defibrillation did not induce ROSC, which was achieved with the recovery of an esophageal temperature of 33.0°C, the additional rapid fall in the patient’s core body temperature during anesthesia may have caused the cardiac arrest.
In this case study, we suspected that restenosis at the site of the previous stent insertion contributed to the cardiac arrest, given the following three findings: First, the patient had a normal preoperative ECG with no signs of ischemia but developed ST-segment depression during anesthesia. Second, repetitive cardiac arrests occurred even after achieving ROSC with rewarming. Third, an increased troponin-T concentration was observed, and suspected re-stenotic occlusion of the left main coronary artery was confirmed by Doppler TTE after the patient’s transfer to the ICU. Accordingly, myocardial infarction was considered the main cause of death.
The mechanisms of myocardial infarction owing to stent restenosis after DES insertion are rapid stent thrombosis or profuse in-stent restenosis (ISR).13 We assume that the lesion characteristics in the left main bifurcation of the coronary artery and DAPT interruption owing to the increasing amount of cerebral hemorrhage contributed to the stent restenosis in this patient. As evidence in support of this assumption, previous studies have shown that stenting at the left main bifurcation and DAPT interruption are strong risk factors for stent thrombosis or ISR.14,15 While controversial, not interrupting aspirin perioperatively could have avoided poor outcomes.16 However, there were no clinical signs related to expected ongoing restenosis of the coronary artery or myocardial ischemia preoperatively in our patient. We suspect that the possible ischemic signs may have been masked by the aforementioned cardioprotective benefits of therapeutic hypothermia. We also believe that the following three conditions contributed to the intraoperative myocardial ischemia and infarction caused by acute restenosis of the coronary stent: First, intraoperative hypotension may have contributed to myocardial injury. The IV anesthetics used during anesthesia, propofol and remifentanil, induced redistribution of blood volume by peripheral vasodilation, resulting in profound hypotension.4 Profound hypotension was strongly associated with myocardial injury and especially harmful for this patient, who had preexisting ischemic heart disease.17 Second, tachycardia, which may have been caused by the inotrope dopamine, used to compensate for the low BP or the stress response to surgical stimulation from shallow anesthetic depth, may have triggered myocardial ischemia by decreasing oxygen delivery and increasing oxygen consumption.17,18 If cardiogenic shock is suspected, as in our case, norepinephrine could be a better option than dopamine.19 Third, the bolus injection of phenylephrine, a selective α1-adrenergic agonist, just before cardiac arrest may have temporarily and abruptly increased cardiac afterload and impaired myocardial perfusion.20
Had we known that coronary artery occlusion would result from hypothermia-induced cardiac arrest and coronary stent restenosis, the treatment of choice would have been CABG surgery while maintaining circulation with emergency cardiopulmonary bypass (CPB) or extracorporeal membrane oxygenation (ECMO) via the femoral artery and vein.11 However, it was difficult to determine the etiology of the cardiac arrest at the time. Even if we had known the cause, ideal treatment would have been impossible, given circumstances beyond our control; all thoracic surgeons were busy operating, and emergency CPB and ECMO were not available. As an alternative, repeat PCI after the patient achieved ROSC could have been performed. However, the patient had a prolonged CPR duration, and he did not maintain the ROSC state for a period sufficient for intervention, which led to a poor outcome.
In this case, it was not possible to accurately determine the reason for the cardiac arrest and for not achieving consistent ROSC. Instead, we conclude that cardiac arrest developed and persisted as a result of the unintentional severe hypothermia that occurred during general anesthesia and/or coronary artery restenosis at the site of the previous PCI. A thorough evaluation of heart-related history is required for TBI patients who have a history of ischemic heart diseases and who undergo therapeutic hypothermia because there may be undiscovered coronary insufficiency owing to the cardioprotective effects of therapeutic hypothermia. In addition, careful monitoring of the patient’s core body temperature, and preventive and countermeasures should be fully prepared and taken against adverse hemodynamic effects, even cardiac arrest, which can occur because of additive hypothermic responses during anesthesia.
Ethics statement: This case report was approved by our Institutional Review Board (EMC 2020-05-005-001), and the case report was performed in compliance with the EQUATOR Network guidelines (the CARE guidelines). All patient details were de-identified, and written consent was obtained from the patient’s guardian for treatment and publication of this report.
Declaration of conflicting interest: The authors declare that there is no conflict of interest.
Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
ORCID iDs: Dong Ho Park https://orcid.org/0000-0002-6587-3756
Tae Woo Kim https://orcid.org/0000-0002-2115-6448
Mo Se Kim https://orcid.org/0000-0001-7362-0030 | DOPAMINE HYDROCHLORIDE, MIDAZOLAM, PHENYLEPHRINE HYDROCHLORIDE, PROPOFOL, REMIFENTANIL, REMIFENTANIL HYDROCHLORIDE, ROCURONIUM BROMIDE, SODIUM CHLORIDE | DrugsGivenReaction | CC BY-NC | 33499678 | 19,429,456 | 2021-01 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Hypocalcaemia'. | Cardiac arrest caused by accidental severe hypothermia and myocardial infarction during general anesthesia.
Therapeutic hypothermia is often used for traumatic brain injury because of its neuroprotective effect and decreased secondary brain injury. However, this procedure lacks clinical evidence supporting its efficacy, and adverse outcomes have been reported during general anesthesia. A 61-year-old man with a history of percutaneous coronary intervention (PCI) was admitted with traumatic brain injury. Immediately after admission, he underwent mild therapeutic hypothermia with a target temperature of 33.0°C for neuroprotection. During general anesthesia for emergency surgery because he developed a mass effect, hypothermic cardiac arrest occurred following an additional decrease in the core body temperature. Moreover, myocardial infarction caused by restenosis of the previous PCI lesion also contributed to the cardiac arrest. Although the patient recovered spontaneous circulation after an hour-long cardiopulmonary resuscitation with rewarming, he eventually died of subsequent repetitive cardiac arrests. When anesthetizing patients undergoing therapeutic hypothermia, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.
Introduction
Therapeutic hypothermia is an attractive strategy for traumatic brain injury (TBI), which demonstrates neuroprotective effects and decreases secondary brain injury.1 Despite the lack of clinical evidence regarding efficacy, therapeutic hypothermia after TBI has been used for at least half a century because of its potential benefits, which are supported by numerous preclinical studies.2,3 However, there have been no reports on the danger of unintentional severe hypothermia during general anesthesia caused by the additional drop in temperature of patients undergoing therapeutic hypothermia. When anesthetizing patients with comorbidities, such as ischemic heart disease, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.4,5
Case report
A 61-year-old man (height, 168 cm; weight, 70 kg) presented to our emergency room (ER) with head trauma after falling on a cement floor from a 5-meter-tall construction site. Upon arrival at the ER, the patient had a Glasgow coma scale score of 11 and the following vital signs: blood pressure (BP), 147/83 mmHg; heart rate (HR), 86 beats per minute (bpm); respiratory rate, 24 breaths per minute; body temperature from the auditory canal, 36.8°C; and percutaneous oxygen saturation (SpO2), 97%. Upon discovering signs of acute subdural hemorrhage, subarachnoid hemorrhage, and intraventricular hemorrhage from a brain computed tomography (CT) scan performed in the ER, we admitted the patient to the intensive care unit (ICU) for close monitoring. After approximately 6 hours, we performed follow-up brain CT because the patient’s mental status had deteriorated from drowsy to stupor. Upon observing a mass effect characterized by an increase in the hemorrhage volume and a left midline shift on the CT images, the patient was scheduled for immediate intubation and emergency decompressive craniectomy (Figure 1).
Figure 1. Brain computed tomography (CT) images taken 6 hours after intensive care unit admission. (a) Axial view showing acute subarachnoid hemorrhage in the peri-mesencephalic cistern (arrow A) and subdural hemorrhage in the right inferior temporal (arrow B) and right parietal (arrow C) regions. (b) Axial view showing acute subdural hemorrhage in the right temporal region (black arrow) and left midline shift.
Although the patient had a history of ischemic heart disease with percutaneous coronary intervention (PCI), it was difficult to obtain accurate information because there was no relevant data in the patient’s medical records from the authors’ hospital, and his drowsy state interfered with detailed interviews. According to the chest CT scan performed immediately after admission, coronary stents had been inserted at the bifurcation site of the left main coronary artery (Figure 2). It also appeared that the inserted stents were drug-eluting stents (DES) considering that he had received dual antiplatelet therapy (DAPT) with aspirin and clopidogrel (Plavix®; Sanofi, Bridgewater, NJ, USA) a switching platelet P2Y12 inhibitor, for approximately 4 months. However, DAPT was discontinued owing to the worsening cerebral hemorrhage after hospitalization; bridging antiplatelet therapy was also not performed. Before the surgery, the patient showed a normal electrocardiogram (ECG), and cardiac biomarker testing revealed a troponin-T concentration of 0.006 µg/L, which was within normal limits (Figure 3). The patient demonstrated satisfactory functional capacity to be able to work at a construction site just before the trauma. Therefore, an additional preoperative cardiac evaluation was not considered necessary.
Figure 2. Chest computed tomography (CT) images showing coronary stents in the main bifurcation of the left coronary arteries (white arrow in (a) and (b)).
Figure 3. Preoperative electrocardiogram of the patient showing normal sinus rhythm with a heart rate of 95 bpm.
bpm, beats per minute.
Immediately after admission to the ICU, the patient’s bladder temperature was 38.0°C. Using a hydrogel-coated water-circulating pad (Arctic Sun®; Medivance Inc., Louisville, CO, USA), which is a surface cooling device, the patient underwent mild therapeutic hypothermia with a target temperature of 33.0°C. During therapeutic hypothermia, sedation and shivering control were performed using continuous rate infusions of midazolam at 5 mg/hour and remifentanil at 0.2 µg/kg/minute. The patient’s bladder temperature was 33.4°C just before he was transferred to the operating room. Preoperative laboratory tests showed no unusual findings except for mildly elevated aspartate aminotransferase (73 U/L) and alanine aminotransferase (51 U/L) concentrations.
The patient’s vital signs upon arrival at the operating room were a BP of 130/70 mmHg and SpO2 of 100%. Lead II of the ECG was monitored, and there were no unusual findings except sinus tachycardia with an HR of 100 bpm on the initial ECG. Arterial BP (ABP), central venous pressure (CVP), and esophageal temperature were monitored from the right radial artery, right subclavian vein, and lower third of the esophagus using a temperature sensor attached to the esophageal stethoscope, respectively. The initial ABP, CVP, and esophageal temperature were 140/85 mmHg, 6 mmHg, and 33.0°C, respectively. Arterial blood gas analysis was performed in the arterial line, with values calculated as for a body temperature of 37°C, which revealed the following: pH, 7.400; partial pressure of carbon dioxide (PCO2), 40 mmHg; partial pressure of oxygen (PO2), 174 mmHg; bicarbonate (HCO3), 24.8 mmol/L; base excess, 0.0; lactate, 0.17 mmol/L; ionized calcium, 1.04 mmol/L; and potassium, 3.3 mmol/L. Total intravenous (IV) anesthesia was performed using a target-controlled infusion of propofol and remifentanil in accordance with the Schnider6 and Minto7 models, respectively. Propofol and remifentanil were infused continuously with effect-site concentrations (Ce) of 1.5 µg/mL and 1.5 ng/mL, respectively. Concurrently, 50 mg of rocuronium was administered by IV bolus injection for neuromuscular blockade. Mechanical ventilation was adjusted to maintain the end-tidal carbon dioxide value at 30 to 33 mmHg.
As ABP decreased to 78/42 mmHg and CVP decreased to 4 mmHg within 10 minutes after anesthesia, the Ce of propofol and remifentanil were lowered to 0.8 µg/mL and 1.0 ng/mL, respectively, and approximately 1 L of balanced crystalloid fluid loading was performed. However, the patient’s BP did not recover, and a dopamine continuous infusion (5 µg/kg/minute) was initiated. When a surgical incision was made 10 minutes after beginning the dopamine infusion, the patient’s ABP and HR increased to 115/70 mmHg and 120 bpm, respectively, and marked ST-segment depression (−2.4 mm) was found, which was not observed on the preoperative or initial ECGs (Figure 4). Subsequently, the depth of anesthesia was increased by increasing the Ce of propofol to 1.0 to 1.2 µg/mL, followed by a continuous infusion of isosorbide dinitrate (1 µg/kg/minute). While HR was maintained at 110 to 120 bpm, ABP gradually decreased to 70/40 mmHg after approximately 5 minutes. To treat the hypotension, isosorbide dinitrate was discontinued, and a bolus injection of 100 µg phenylephrine was administered, after which, ABP increased temporarily to 95/60 mmHg. Subsequently, HR decreased suddenly and rapidly, and ventricular fibrillation occurred. The anesthesiologists immediately discontinued the surgery and performed cardiopulmonary resuscitation (CPR). The patient’s esophageal temperature at the time was 32.0°C.
Figure 4. Intraoperative electrocardiogram of the patient showing sinus tachycardia with significant ST depression. The ST depression on the monitor represented approximately −2.4 at the time.
Chest compressions were performed with periodic defibrillation with 200 J, and epinephrine injections were administered every 3 minutes; these measures elicited no response. During CPR, ECG monitoring showed temporary severe bradycardia (HR, 10–30 bpm). However, this was considered pulseless electrical activity because no pulse was detected in the patient’s femoral artery. Soon afterward, there was a rapid transition to ventricular fibrillation, and CPR was repeated. At the time, the patient’s esophageal temperature dropped to 30.0°C. According to the refractory state after the defibrillation and drug injections, we considered that the cardiac arrest was caused by hypothermia. To reverse the cause, a forced-air warming device and a warming blanket (Bair Hugger®; Arizant Healthcare Inc., Eden Prairie, MN, USA) were used for surface warming, the IV fluids were warmed, and surgical wound closure and dressing were performed rapidly to prevent surface heat loss. The first arterial blood gas analysis after ventricular fibrillation, calculated as for a body temperature of 37°C, showed the following: pH, 7.690; PCO2, 15 mmHg; PO2, 260 mmHg; HCO3, 18.1 mmol/L; base excess, −1.2; lactate, 0.33 mmol/L; ionized calcium, 0.88 mmol/L; and potassium, 5.0 mmol/L. Follow-up arterial blood gas analyses were repeated periodically. As we continued to perform CPR, the patient developed severe metabolic acidosis and hypocalcemia characterized by a pH of 6.840, HCO3 of 15.4 mmol/L, and ionized calcium of 0.66 mmol/L. To correct the abnormal findings, we administered appropriate amounts of sodium bicarbonate and calcium chloride. After an hour-long CPR and rewarming, the patient’s esophageal temperature recovered to 33.0°C, which was followed by observing sinus rhythm characterized by an HR of 75 bpm on an ECG monitor and detecting a pulse in the femoral artery with a characteristic ABP waveform that eventually led to the recovery of BP to 130/55 mmHg. Upon confirming recovery of spontaneous circulation (ROSC), the patient was transported to the ICU while we administered dopamine (10 µg/kg/minute), norepinephrine (0.1 µg/kg/minute), and vasopressin (2 U/hour). ABP, HR, and esophageal temperature during anesthesia are shown in Figure 5.
Figure 5. Changes in vital signs (blood pressure (BP), heart rate (HR), and esophageal temperature) over time (5-minute intervals).
bpm, beats per minute.
Cardiac biomarker testing performed after the patient’s transfer to the ICU showed an increased troponin-T level of 1.432 µg/L. Portable Doppler transthoracic echocardiography (TTE) revealed total occlusion of the left main coronary artery, which had undergone PCI previously. Because emergency coronary artery bypass grafting (CABG) was impossible at the time, repeated PCI was considered an alternative treatment option. However, even after his transfer to the ICU, the patient experienced repetitive ventricular fibrillation. Four hours after the patient’s transfer to the ICU, the patient’s guardian requested that CPR should be stopped. Without resuscitation, the patient died.
Discussion
In this case, the patient, who was receiving therapeutic hypothermia, underwent emergency craniectomy owing to worsening brain hemorrhage, and cardiac arrest occurred during general anesthesia. We deduced that the cardiac arrest was owing to the additional decrease in the patient’s core body temperature and myocardial infarction caused by restenosis of the previous PCI-treated lesion.
Therapeutic hypothermia is used to improve neurological outcomes in cardiac arrest, TBI, stroke, and hypoxic–ischemic encephalopathy.1,4 Studies on the efficacy of therapeutic hypothermia for TBI have been less convincing, and they show conflicting results. However, hypothermia is consistently used as an option for TBI because it reduces intracranial pressure and the cerebral metabolic rate, thereby protecting neuronal function in the brain and decreasing the wide range of injuries or reparative processes by attenuating temperature-sensitive mechanisms, such as excitotoxicity, free radical generation, apoptosis, and inflammation.1,2
The cardiovascular effects associated with hypothermia are complex. Hypothermia affects the electrical conduction of myocardial cells, which can induce hypotension or bradycardia and even increase the risk of ventricular fibrillation by prolonging repolarization.4,8 However, mild hypothermia of 32.0°C to 35.0°C demonstrates cardioprotective effects by decreasing myocardial oxygen demand and increasing myocardial perfusion through vasodilation of coronary vessels.9 Therefore, there has been a growing interest in the clinical use of mild therapeutic hypothermia as an adjunctive therapy during ischemic heart disease.10 Our patient received mild therapeutic hypothermia with a target core temperature of 33.0°C, which may have yielded cardioprotective as well as neuroprotective benefits.11
During general anesthesia, hypothermia is induced following an increase in heat loss caused by peripheral vasodilation after the injection of anesthetic drugs, heat redistribution from the core to the periphery, and impaired normal thermoregulatory responses, such as vasoconstriction or shivering.5 In our patient, the cold operating room temperature and irrigating an uncovered surgical area in the brain with cold saline also may have contributed to the additional core body temperature drop in addition to the effects of general anesthesia.12 However, in our case, neurosurgeons requested that anesthesiologists not perform warming during anesthesia. Instead, esophageal temperature and signs that could result from hypothermia, such as ECG changes, were carefully monitored. The esophageal temperature decreased from 33.0°C, which was measured immediately after entering the operation room, to 32.0°C at the time of cardiac arrest. The esophageal temperature further decreased to a minimum of 30.0°C during CPR. With rewarming, the temperature recovered to 33.0°C and finally, ROSC was achieved. Considering that chest compressions, drugs, and defibrillation did not induce ROSC, which was achieved with the recovery of an esophageal temperature of 33.0°C, the additional rapid fall in the patient’s core body temperature during anesthesia may have caused the cardiac arrest.
In this case study, we suspected that restenosis at the site of the previous stent insertion contributed to the cardiac arrest, given the following three findings: First, the patient had a normal preoperative ECG with no signs of ischemia but developed ST-segment depression during anesthesia. Second, repetitive cardiac arrests occurred even after achieving ROSC with rewarming. Third, an increased troponin-T concentration was observed, and suspected re-stenotic occlusion of the left main coronary artery was confirmed by Doppler TTE after the patient’s transfer to the ICU. Accordingly, myocardial infarction was considered the main cause of death.
The mechanisms of myocardial infarction owing to stent restenosis after DES insertion are rapid stent thrombosis or profuse in-stent restenosis (ISR).13 We assume that the lesion characteristics in the left main bifurcation of the coronary artery and DAPT interruption owing to the increasing amount of cerebral hemorrhage contributed to the stent restenosis in this patient. As evidence in support of this assumption, previous studies have shown that stenting at the left main bifurcation and DAPT interruption are strong risk factors for stent thrombosis or ISR.14,15 While controversial, not interrupting aspirin perioperatively could have avoided poor outcomes.16 However, there were no clinical signs related to expected ongoing restenosis of the coronary artery or myocardial ischemia preoperatively in our patient. We suspect that the possible ischemic signs may have been masked by the aforementioned cardioprotective benefits of therapeutic hypothermia. We also believe that the following three conditions contributed to the intraoperative myocardial ischemia and infarction caused by acute restenosis of the coronary stent: First, intraoperative hypotension may have contributed to myocardial injury. The IV anesthetics used during anesthesia, propofol and remifentanil, induced redistribution of blood volume by peripheral vasodilation, resulting in profound hypotension.4 Profound hypotension was strongly associated with myocardial injury and especially harmful for this patient, who had preexisting ischemic heart disease.17 Second, tachycardia, which may have been caused by the inotrope dopamine, used to compensate for the low BP or the stress response to surgical stimulation from shallow anesthetic depth, may have triggered myocardial ischemia by decreasing oxygen delivery and increasing oxygen consumption.17,18 If cardiogenic shock is suspected, as in our case, norepinephrine could be a better option than dopamine.19 Third, the bolus injection of phenylephrine, a selective α1-adrenergic agonist, just before cardiac arrest may have temporarily and abruptly increased cardiac afterload and impaired myocardial perfusion.20
Had we known that coronary artery occlusion would result from hypothermia-induced cardiac arrest and coronary stent restenosis, the treatment of choice would have been CABG surgery while maintaining circulation with emergency cardiopulmonary bypass (CPB) or extracorporeal membrane oxygenation (ECMO) via the femoral artery and vein.11 However, it was difficult to determine the etiology of the cardiac arrest at the time. Even if we had known the cause, ideal treatment would have been impossible, given circumstances beyond our control; all thoracic surgeons were busy operating, and emergency CPB and ECMO were not available. As an alternative, repeat PCI after the patient achieved ROSC could have been performed. However, the patient had a prolonged CPR duration, and he did not maintain the ROSC state for a period sufficient for intervention, which led to a poor outcome.
In this case, it was not possible to accurately determine the reason for the cardiac arrest and for not achieving consistent ROSC. Instead, we conclude that cardiac arrest developed and persisted as a result of the unintentional severe hypothermia that occurred during general anesthesia and/or coronary artery restenosis at the site of the previous PCI. A thorough evaluation of heart-related history is required for TBI patients who have a history of ischemic heart diseases and who undergo therapeutic hypothermia because there may be undiscovered coronary insufficiency owing to the cardioprotective effects of therapeutic hypothermia. In addition, careful monitoring of the patient’s core body temperature, and preventive and countermeasures should be fully prepared and taken against adverse hemodynamic effects, even cardiac arrest, which can occur because of additive hypothermic responses during anesthesia.
Ethics statement: This case report was approved by our Institutional Review Board (EMC 2020-05-005-001), and the case report was performed in compliance with the EQUATOR Network guidelines (the CARE guidelines). All patient details were de-identified, and written consent was obtained from the patient’s guardian for treatment and publication of this report.
Declaration of conflicting interest: The authors declare that there is no conflict of interest.
Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
ORCID iDs: Dong Ho Park https://orcid.org/0000-0002-6587-3756
Tae Woo Kim https://orcid.org/0000-0002-2115-6448
Mo Se Kim https://orcid.org/0000-0001-7362-0030 | DOPAMINE HYDROCHLORIDE, MIDAZOLAM, PHENYLEPHRINE HYDROCHLORIDE, PROPOFOL, REMIFENTANIL, REMIFENTANIL HYDROCHLORIDE, ROCURONIUM BROMIDE, SODIUM CHLORIDE | DrugsGivenReaction | CC BY-NC | 33499678 | 19,429,456 | 2021-01 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Hypothermia'. | Cardiac arrest caused by accidental severe hypothermia and myocardial infarction during general anesthesia.
Therapeutic hypothermia is often used for traumatic brain injury because of its neuroprotective effect and decreased secondary brain injury. However, this procedure lacks clinical evidence supporting its efficacy, and adverse outcomes have been reported during general anesthesia. A 61-year-old man with a history of percutaneous coronary intervention (PCI) was admitted with traumatic brain injury. Immediately after admission, he underwent mild therapeutic hypothermia with a target temperature of 33.0°C for neuroprotection. During general anesthesia for emergency surgery because he developed a mass effect, hypothermic cardiac arrest occurred following an additional decrease in the core body temperature. Moreover, myocardial infarction caused by restenosis of the previous PCI lesion also contributed to the cardiac arrest. Although the patient recovered spontaneous circulation after an hour-long cardiopulmonary resuscitation with rewarming, he eventually died of subsequent repetitive cardiac arrests. When anesthetizing patients undergoing therapeutic hypothermia, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.
Introduction
Therapeutic hypothermia is an attractive strategy for traumatic brain injury (TBI), which demonstrates neuroprotective effects and decreases secondary brain injury.1 Despite the lack of clinical evidence regarding efficacy, therapeutic hypothermia after TBI has been used for at least half a century because of its potential benefits, which are supported by numerous preclinical studies.2,3 However, there have been no reports on the danger of unintentional severe hypothermia during general anesthesia caused by the additional drop in temperature of patients undergoing therapeutic hypothermia. When anesthetizing patients with comorbidities, such as ischemic heart disease, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.4,5
Case report
A 61-year-old man (height, 168 cm; weight, 70 kg) presented to our emergency room (ER) with head trauma after falling on a cement floor from a 5-meter-tall construction site. Upon arrival at the ER, the patient had a Glasgow coma scale score of 11 and the following vital signs: blood pressure (BP), 147/83 mmHg; heart rate (HR), 86 beats per minute (bpm); respiratory rate, 24 breaths per minute; body temperature from the auditory canal, 36.8°C; and percutaneous oxygen saturation (SpO2), 97%. Upon discovering signs of acute subdural hemorrhage, subarachnoid hemorrhage, and intraventricular hemorrhage from a brain computed tomography (CT) scan performed in the ER, we admitted the patient to the intensive care unit (ICU) for close monitoring. After approximately 6 hours, we performed follow-up brain CT because the patient’s mental status had deteriorated from drowsy to stupor. Upon observing a mass effect characterized by an increase in the hemorrhage volume and a left midline shift on the CT images, the patient was scheduled for immediate intubation and emergency decompressive craniectomy (Figure 1).
Figure 1. Brain computed tomography (CT) images taken 6 hours after intensive care unit admission. (a) Axial view showing acute subarachnoid hemorrhage in the peri-mesencephalic cistern (arrow A) and subdural hemorrhage in the right inferior temporal (arrow B) and right parietal (arrow C) regions. (b) Axial view showing acute subdural hemorrhage in the right temporal region (black arrow) and left midline shift.
Although the patient had a history of ischemic heart disease with percutaneous coronary intervention (PCI), it was difficult to obtain accurate information because there was no relevant data in the patient’s medical records from the authors’ hospital, and his drowsy state interfered with detailed interviews. According to the chest CT scan performed immediately after admission, coronary stents had been inserted at the bifurcation site of the left main coronary artery (Figure 2). It also appeared that the inserted stents were drug-eluting stents (DES) considering that he had received dual antiplatelet therapy (DAPT) with aspirin and clopidogrel (Plavix®; Sanofi, Bridgewater, NJ, USA) a switching platelet P2Y12 inhibitor, for approximately 4 months. However, DAPT was discontinued owing to the worsening cerebral hemorrhage after hospitalization; bridging antiplatelet therapy was also not performed. Before the surgery, the patient showed a normal electrocardiogram (ECG), and cardiac biomarker testing revealed a troponin-T concentration of 0.006 µg/L, which was within normal limits (Figure 3). The patient demonstrated satisfactory functional capacity to be able to work at a construction site just before the trauma. Therefore, an additional preoperative cardiac evaluation was not considered necessary.
Figure 2. Chest computed tomography (CT) images showing coronary stents in the main bifurcation of the left coronary arteries (white arrow in (a) and (b)).
Figure 3. Preoperative electrocardiogram of the patient showing normal sinus rhythm with a heart rate of 95 bpm.
bpm, beats per minute.
Immediately after admission to the ICU, the patient’s bladder temperature was 38.0°C. Using a hydrogel-coated water-circulating pad (Arctic Sun®; Medivance Inc., Louisville, CO, USA), which is a surface cooling device, the patient underwent mild therapeutic hypothermia with a target temperature of 33.0°C. During therapeutic hypothermia, sedation and shivering control were performed using continuous rate infusions of midazolam at 5 mg/hour and remifentanil at 0.2 µg/kg/minute. The patient’s bladder temperature was 33.4°C just before he was transferred to the operating room. Preoperative laboratory tests showed no unusual findings except for mildly elevated aspartate aminotransferase (73 U/L) and alanine aminotransferase (51 U/L) concentrations.
The patient’s vital signs upon arrival at the operating room were a BP of 130/70 mmHg and SpO2 of 100%. Lead II of the ECG was monitored, and there were no unusual findings except sinus tachycardia with an HR of 100 bpm on the initial ECG. Arterial BP (ABP), central venous pressure (CVP), and esophageal temperature were monitored from the right radial artery, right subclavian vein, and lower third of the esophagus using a temperature sensor attached to the esophageal stethoscope, respectively. The initial ABP, CVP, and esophageal temperature were 140/85 mmHg, 6 mmHg, and 33.0°C, respectively. Arterial blood gas analysis was performed in the arterial line, with values calculated as for a body temperature of 37°C, which revealed the following: pH, 7.400; partial pressure of carbon dioxide (PCO2), 40 mmHg; partial pressure of oxygen (PO2), 174 mmHg; bicarbonate (HCO3), 24.8 mmol/L; base excess, 0.0; lactate, 0.17 mmol/L; ionized calcium, 1.04 mmol/L; and potassium, 3.3 mmol/L. Total intravenous (IV) anesthesia was performed using a target-controlled infusion of propofol and remifentanil in accordance with the Schnider6 and Minto7 models, respectively. Propofol and remifentanil were infused continuously with effect-site concentrations (Ce) of 1.5 µg/mL and 1.5 ng/mL, respectively. Concurrently, 50 mg of rocuronium was administered by IV bolus injection for neuromuscular blockade. Mechanical ventilation was adjusted to maintain the end-tidal carbon dioxide value at 30 to 33 mmHg.
As ABP decreased to 78/42 mmHg and CVP decreased to 4 mmHg within 10 minutes after anesthesia, the Ce of propofol and remifentanil were lowered to 0.8 µg/mL and 1.0 ng/mL, respectively, and approximately 1 L of balanced crystalloid fluid loading was performed. However, the patient’s BP did not recover, and a dopamine continuous infusion (5 µg/kg/minute) was initiated. When a surgical incision was made 10 minutes after beginning the dopamine infusion, the patient’s ABP and HR increased to 115/70 mmHg and 120 bpm, respectively, and marked ST-segment depression (−2.4 mm) was found, which was not observed on the preoperative or initial ECGs (Figure 4). Subsequently, the depth of anesthesia was increased by increasing the Ce of propofol to 1.0 to 1.2 µg/mL, followed by a continuous infusion of isosorbide dinitrate (1 µg/kg/minute). While HR was maintained at 110 to 120 bpm, ABP gradually decreased to 70/40 mmHg after approximately 5 minutes. To treat the hypotension, isosorbide dinitrate was discontinued, and a bolus injection of 100 µg phenylephrine was administered, after which, ABP increased temporarily to 95/60 mmHg. Subsequently, HR decreased suddenly and rapidly, and ventricular fibrillation occurred. The anesthesiologists immediately discontinued the surgery and performed cardiopulmonary resuscitation (CPR). The patient’s esophageal temperature at the time was 32.0°C.
Figure 4. Intraoperative electrocardiogram of the patient showing sinus tachycardia with significant ST depression. The ST depression on the monitor represented approximately −2.4 at the time.
Chest compressions were performed with periodic defibrillation with 200 J, and epinephrine injections were administered every 3 minutes; these measures elicited no response. During CPR, ECG monitoring showed temporary severe bradycardia (HR, 10–30 bpm). However, this was considered pulseless electrical activity because no pulse was detected in the patient’s femoral artery. Soon afterward, there was a rapid transition to ventricular fibrillation, and CPR was repeated. At the time, the patient’s esophageal temperature dropped to 30.0°C. According to the refractory state after the defibrillation and drug injections, we considered that the cardiac arrest was caused by hypothermia. To reverse the cause, a forced-air warming device and a warming blanket (Bair Hugger®; Arizant Healthcare Inc., Eden Prairie, MN, USA) were used for surface warming, the IV fluids were warmed, and surgical wound closure and dressing were performed rapidly to prevent surface heat loss. The first arterial blood gas analysis after ventricular fibrillation, calculated as for a body temperature of 37°C, showed the following: pH, 7.690; PCO2, 15 mmHg; PO2, 260 mmHg; HCO3, 18.1 mmol/L; base excess, −1.2; lactate, 0.33 mmol/L; ionized calcium, 0.88 mmol/L; and potassium, 5.0 mmol/L. Follow-up arterial blood gas analyses were repeated periodically. As we continued to perform CPR, the patient developed severe metabolic acidosis and hypocalcemia characterized by a pH of 6.840, HCO3 of 15.4 mmol/L, and ionized calcium of 0.66 mmol/L. To correct the abnormal findings, we administered appropriate amounts of sodium bicarbonate and calcium chloride. After an hour-long CPR and rewarming, the patient’s esophageal temperature recovered to 33.0°C, which was followed by observing sinus rhythm characterized by an HR of 75 bpm on an ECG monitor and detecting a pulse in the femoral artery with a characteristic ABP waveform that eventually led to the recovery of BP to 130/55 mmHg. Upon confirming recovery of spontaneous circulation (ROSC), the patient was transported to the ICU while we administered dopamine (10 µg/kg/minute), norepinephrine (0.1 µg/kg/minute), and vasopressin (2 U/hour). ABP, HR, and esophageal temperature during anesthesia are shown in Figure 5.
Figure 5. Changes in vital signs (blood pressure (BP), heart rate (HR), and esophageal temperature) over time (5-minute intervals).
bpm, beats per minute.
Cardiac biomarker testing performed after the patient’s transfer to the ICU showed an increased troponin-T level of 1.432 µg/L. Portable Doppler transthoracic echocardiography (TTE) revealed total occlusion of the left main coronary artery, which had undergone PCI previously. Because emergency coronary artery bypass grafting (CABG) was impossible at the time, repeated PCI was considered an alternative treatment option. However, even after his transfer to the ICU, the patient experienced repetitive ventricular fibrillation. Four hours after the patient’s transfer to the ICU, the patient’s guardian requested that CPR should be stopped. Without resuscitation, the patient died.
Discussion
In this case, the patient, who was receiving therapeutic hypothermia, underwent emergency craniectomy owing to worsening brain hemorrhage, and cardiac arrest occurred during general anesthesia. We deduced that the cardiac arrest was owing to the additional decrease in the patient’s core body temperature and myocardial infarction caused by restenosis of the previous PCI-treated lesion.
Therapeutic hypothermia is used to improve neurological outcomes in cardiac arrest, TBI, stroke, and hypoxic–ischemic encephalopathy.1,4 Studies on the efficacy of therapeutic hypothermia for TBI have been less convincing, and they show conflicting results. However, hypothermia is consistently used as an option for TBI because it reduces intracranial pressure and the cerebral metabolic rate, thereby protecting neuronal function in the brain and decreasing the wide range of injuries or reparative processes by attenuating temperature-sensitive mechanisms, such as excitotoxicity, free radical generation, apoptosis, and inflammation.1,2
The cardiovascular effects associated with hypothermia are complex. Hypothermia affects the electrical conduction of myocardial cells, which can induce hypotension or bradycardia and even increase the risk of ventricular fibrillation by prolonging repolarization.4,8 However, mild hypothermia of 32.0°C to 35.0°C demonstrates cardioprotective effects by decreasing myocardial oxygen demand and increasing myocardial perfusion through vasodilation of coronary vessels.9 Therefore, there has been a growing interest in the clinical use of mild therapeutic hypothermia as an adjunctive therapy during ischemic heart disease.10 Our patient received mild therapeutic hypothermia with a target core temperature of 33.0°C, which may have yielded cardioprotective as well as neuroprotective benefits.11
During general anesthesia, hypothermia is induced following an increase in heat loss caused by peripheral vasodilation after the injection of anesthetic drugs, heat redistribution from the core to the periphery, and impaired normal thermoregulatory responses, such as vasoconstriction or shivering.5 In our patient, the cold operating room temperature and irrigating an uncovered surgical area in the brain with cold saline also may have contributed to the additional core body temperature drop in addition to the effects of general anesthesia.12 However, in our case, neurosurgeons requested that anesthesiologists not perform warming during anesthesia. Instead, esophageal temperature and signs that could result from hypothermia, such as ECG changes, were carefully monitored. The esophageal temperature decreased from 33.0°C, which was measured immediately after entering the operation room, to 32.0°C at the time of cardiac arrest. The esophageal temperature further decreased to a minimum of 30.0°C during CPR. With rewarming, the temperature recovered to 33.0°C and finally, ROSC was achieved. Considering that chest compressions, drugs, and defibrillation did not induce ROSC, which was achieved with the recovery of an esophageal temperature of 33.0°C, the additional rapid fall in the patient’s core body temperature during anesthesia may have caused the cardiac arrest.
In this case study, we suspected that restenosis at the site of the previous stent insertion contributed to the cardiac arrest, given the following three findings: First, the patient had a normal preoperative ECG with no signs of ischemia but developed ST-segment depression during anesthesia. Second, repetitive cardiac arrests occurred even after achieving ROSC with rewarming. Third, an increased troponin-T concentration was observed, and suspected re-stenotic occlusion of the left main coronary artery was confirmed by Doppler TTE after the patient’s transfer to the ICU. Accordingly, myocardial infarction was considered the main cause of death.
The mechanisms of myocardial infarction owing to stent restenosis after DES insertion are rapid stent thrombosis or profuse in-stent restenosis (ISR).13 We assume that the lesion characteristics in the left main bifurcation of the coronary artery and DAPT interruption owing to the increasing amount of cerebral hemorrhage contributed to the stent restenosis in this patient. As evidence in support of this assumption, previous studies have shown that stenting at the left main bifurcation and DAPT interruption are strong risk factors for stent thrombosis or ISR.14,15 While controversial, not interrupting aspirin perioperatively could have avoided poor outcomes.16 However, there were no clinical signs related to expected ongoing restenosis of the coronary artery or myocardial ischemia preoperatively in our patient. We suspect that the possible ischemic signs may have been masked by the aforementioned cardioprotective benefits of therapeutic hypothermia. We also believe that the following three conditions contributed to the intraoperative myocardial ischemia and infarction caused by acute restenosis of the coronary stent: First, intraoperative hypotension may have contributed to myocardial injury. The IV anesthetics used during anesthesia, propofol and remifentanil, induced redistribution of blood volume by peripheral vasodilation, resulting in profound hypotension.4 Profound hypotension was strongly associated with myocardial injury and especially harmful for this patient, who had preexisting ischemic heart disease.17 Second, tachycardia, which may have been caused by the inotrope dopamine, used to compensate for the low BP or the stress response to surgical stimulation from shallow anesthetic depth, may have triggered myocardial ischemia by decreasing oxygen delivery and increasing oxygen consumption.17,18 If cardiogenic shock is suspected, as in our case, norepinephrine could be a better option than dopamine.19 Third, the bolus injection of phenylephrine, a selective α1-adrenergic agonist, just before cardiac arrest may have temporarily and abruptly increased cardiac afterload and impaired myocardial perfusion.20
Had we known that coronary artery occlusion would result from hypothermia-induced cardiac arrest and coronary stent restenosis, the treatment of choice would have been CABG surgery while maintaining circulation with emergency cardiopulmonary bypass (CPB) or extracorporeal membrane oxygenation (ECMO) via the femoral artery and vein.11 However, it was difficult to determine the etiology of the cardiac arrest at the time. Even if we had known the cause, ideal treatment would have been impossible, given circumstances beyond our control; all thoracic surgeons were busy operating, and emergency CPB and ECMO were not available. As an alternative, repeat PCI after the patient achieved ROSC could have been performed. However, the patient had a prolonged CPR duration, and he did not maintain the ROSC state for a period sufficient for intervention, which led to a poor outcome.
In this case, it was not possible to accurately determine the reason for the cardiac arrest and for not achieving consistent ROSC. Instead, we conclude that cardiac arrest developed and persisted as a result of the unintentional severe hypothermia that occurred during general anesthesia and/or coronary artery restenosis at the site of the previous PCI. A thorough evaluation of heart-related history is required for TBI patients who have a history of ischemic heart diseases and who undergo therapeutic hypothermia because there may be undiscovered coronary insufficiency owing to the cardioprotective effects of therapeutic hypothermia. In addition, careful monitoring of the patient’s core body temperature, and preventive and countermeasures should be fully prepared and taken against adverse hemodynamic effects, even cardiac arrest, which can occur because of additive hypothermic responses during anesthesia.
Ethics statement: This case report was approved by our Institutional Review Board (EMC 2020-05-005-001), and the case report was performed in compliance with the EQUATOR Network guidelines (the CARE guidelines). All patient details were de-identified, and written consent was obtained from the patient’s guardian for treatment and publication of this report.
Declaration of conflicting interest: The authors declare that there is no conflict of interest.
Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
ORCID iDs: Dong Ho Park https://orcid.org/0000-0002-6587-3756
Tae Woo Kim https://orcid.org/0000-0002-2115-6448
Mo Se Kim https://orcid.org/0000-0001-7362-0030 | DOPAMINE HYDROCHLORIDE, MIDAZOLAM, PHENYLEPHRINE HYDROCHLORIDE, PROPOFOL, REMIFENTANIL, REMIFENTANIL HYDROCHLORIDE, ROCURONIUM BROMIDE, SODIUM CHLORIDE | DrugsGivenReaction | CC BY-NC | 33499678 | 19,429,456 | 2021-01 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Ischaemia'. | Cardiac arrest caused by accidental severe hypothermia and myocardial infarction during general anesthesia.
Therapeutic hypothermia is often used for traumatic brain injury because of its neuroprotective effect and decreased secondary brain injury. However, this procedure lacks clinical evidence supporting its efficacy, and adverse outcomes have been reported during general anesthesia. A 61-year-old man with a history of percutaneous coronary intervention (PCI) was admitted with traumatic brain injury. Immediately after admission, he underwent mild therapeutic hypothermia with a target temperature of 33.0°C for neuroprotection. During general anesthesia for emergency surgery because he developed a mass effect, hypothermic cardiac arrest occurred following an additional decrease in the core body temperature. Moreover, myocardial infarction caused by restenosis of the previous PCI lesion also contributed to the cardiac arrest. Although the patient recovered spontaneous circulation after an hour-long cardiopulmonary resuscitation with rewarming, he eventually died of subsequent repetitive cardiac arrests. When anesthetizing patients undergoing therapeutic hypothermia, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.
Introduction
Therapeutic hypothermia is an attractive strategy for traumatic brain injury (TBI), which demonstrates neuroprotective effects and decreases secondary brain injury.1 Despite the lack of clinical evidence regarding efficacy, therapeutic hypothermia after TBI has been used for at least half a century because of its potential benefits, which are supported by numerous preclinical studies.2,3 However, there have been no reports on the danger of unintentional severe hypothermia during general anesthesia caused by the additional drop in temperature of patients undergoing therapeutic hypothermia. When anesthetizing patients with comorbidities, such as ischemic heart disease, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.4,5
Case report
A 61-year-old man (height, 168 cm; weight, 70 kg) presented to our emergency room (ER) with head trauma after falling on a cement floor from a 5-meter-tall construction site. Upon arrival at the ER, the patient had a Glasgow coma scale score of 11 and the following vital signs: blood pressure (BP), 147/83 mmHg; heart rate (HR), 86 beats per minute (bpm); respiratory rate, 24 breaths per minute; body temperature from the auditory canal, 36.8°C; and percutaneous oxygen saturation (SpO2), 97%. Upon discovering signs of acute subdural hemorrhage, subarachnoid hemorrhage, and intraventricular hemorrhage from a brain computed tomography (CT) scan performed in the ER, we admitted the patient to the intensive care unit (ICU) for close monitoring. After approximately 6 hours, we performed follow-up brain CT because the patient’s mental status had deteriorated from drowsy to stupor. Upon observing a mass effect characterized by an increase in the hemorrhage volume and a left midline shift on the CT images, the patient was scheduled for immediate intubation and emergency decompressive craniectomy (Figure 1).
Figure 1. Brain computed tomography (CT) images taken 6 hours after intensive care unit admission. (a) Axial view showing acute subarachnoid hemorrhage in the peri-mesencephalic cistern (arrow A) and subdural hemorrhage in the right inferior temporal (arrow B) and right parietal (arrow C) regions. (b) Axial view showing acute subdural hemorrhage in the right temporal region (black arrow) and left midline shift.
Although the patient had a history of ischemic heart disease with percutaneous coronary intervention (PCI), it was difficult to obtain accurate information because there was no relevant data in the patient’s medical records from the authors’ hospital, and his drowsy state interfered with detailed interviews. According to the chest CT scan performed immediately after admission, coronary stents had been inserted at the bifurcation site of the left main coronary artery (Figure 2). It also appeared that the inserted stents were drug-eluting stents (DES) considering that he had received dual antiplatelet therapy (DAPT) with aspirin and clopidogrel (Plavix®; Sanofi, Bridgewater, NJ, USA) a switching platelet P2Y12 inhibitor, for approximately 4 months. However, DAPT was discontinued owing to the worsening cerebral hemorrhage after hospitalization; bridging antiplatelet therapy was also not performed. Before the surgery, the patient showed a normal electrocardiogram (ECG), and cardiac biomarker testing revealed a troponin-T concentration of 0.006 µg/L, which was within normal limits (Figure 3). The patient demonstrated satisfactory functional capacity to be able to work at a construction site just before the trauma. Therefore, an additional preoperative cardiac evaluation was not considered necessary.
Figure 2. Chest computed tomography (CT) images showing coronary stents in the main bifurcation of the left coronary arteries (white arrow in (a) and (b)).
Figure 3. Preoperative electrocardiogram of the patient showing normal sinus rhythm with a heart rate of 95 bpm.
bpm, beats per minute.
Immediately after admission to the ICU, the patient’s bladder temperature was 38.0°C. Using a hydrogel-coated water-circulating pad (Arctic Sun®; Medivance Inc., Louisville, CO, USA), which is a surface cooling device, the patient underwent mild therapeutic hypothermia with a target temperature of 33.0°C. During therapeutic hypothermia, sedation and shivering control were performed using continuous rate infusions of midazolam at 5 mg/hour and remifentanil at 0.2 µg/kg/minute. The patient’s bladder temperature was 33.4°C just before he was transferred to the operating room. Preoperative laboratory tests showed no unusual findings except for mildly elevated aspartate aminotransferase (73 U/L) and alanine aminotransferase (51 U/L) concentrations.
The patient’s vital signs upon arrival at the operating room were a BP of 130/70 mmHg and SpO2 of 100%. Lead II of the ECG was monitored, and there were no unusual findings except sinus tachycardia with an HR of 100 bpm on the initial ECG. Arterial BP (ABP), central venous pressure (CVP), and esophageal temperature were monitored from the right radial artery, right subclavian vein, and lower third of the esophagus using a temperature sensor attached to the esophageal stethoscope, respectively. The initial ABP, CVP, and esophageal temperature were 140/85 mmHg, 6 mmHg, and 33.0°C, respectively. Arterial blood gas analysis was performed in the arterial line, with values calculated as for a body temperature of 37°C, which revealed the following: pH, 7.400; partial pressure of carbon dioxide (PCO2), 40 mmHg; partial pressure of oxygen (PO2), 174 mmHg; bicarbonate (HCO3), 24.8 mmol/L; base excess, 0.0; lactate, 0.17 mmol/L; ionized calcium, 1.04 mmol/L; and potassium, 3.3 mmol/L. Total intravenous (IV) anesthesia was performed using a target-controlled infusion of propofol and remifentanil in accordance with the Schnider6 and Minto7 models, respectively. Propofol and remifentanil were infused continuously with effect-site concentrations (Ce) of 1.5 µg/mL and 1.5 ng/mL, respectively. Concurrently, 50 mg of rocuronium was administered by IV bolus injection for neuromuscular blockade. Mechanical ventilation was adjusted to maintain the end-tidal carbon dioxide value at 30 to 33 mmHg.
As ABP decreased to 78/42 mmHg and CVP decreased to 4 mmHg within 10 minutes after anesthesia, the Ce of propofol and remifentanil were lowered to 0.8 µg/mL and 1.0 ng/mL, respectively, and approximately 1 L of balanced crystalloid fluid loading was performed. However, the patient’s BP did not recover, and a dopamine continuous infusion (5 µg/kg/minute) was initiated. When a surgical incision was made 10 minutes after beginning the dopamine infusion, the patient’s ABP and HR increased to 115/70 mmHg and 120 bpm, respectively, and marked ST-segment depression (−2.4 mm) was found, which was not observed on the preoperative or initial ECGs (Figure 4). Subsequently, the depth of anesthesia was increased by increasing the Ce of propofol to 1.0 to 1.2 µg/mL, followed by a continuous infusion of isosorbide dinitrate (1 µg/kg/minute). While HR was maintained at 110 to 120 bpm, ABP gradually decreased to 70/40 mmHg after approximately 5 minutes. To treat the hypotension, isosorbide dinitrate was discontinued, and a bolus injection of 100 µg phenylephrine was administered, after which, ABP increased temporarily to 95/60 mmHg. Subsequently, HR decreased suddenly and rapidly, and ventricular fibrillation occurred. The anesthesiologists immediately discontinued the surgery and performed cardiopulmonary resuscitation (CPR). The patient’s esophageal temperature at the time was 32.0°C.
Figure 4. Intraoperative electrocardiogram of the patient showing sinus tachycardia with significant ST depression. The ST depression on the monitor represented approximately −2.4 at the time.
Chest compressions were performed with periodic defibrillation with 200 J, and epinephrine injections were administered every 3 minutes; these measures elicited no response. During CPR, ECG monitoring showed temporary severe bradycardia (HR, 10–30 bpm). However, this was considered pulseless electrical activity because no pulse was detected in the patient’s femoral artery. Soon afterward, there was a rapid transition to ventricular fibrillation, and CPR was repeated. At the time, the patient’s esophageal temperature dropped to 30.0°C. According to the refractory state after the defibrillation and drug injections, we considered that the cardiac arrest was caused by hypothermia. To reverse the cause, a forced-air warming device and a warming blanket (Bair Hugger®; Arizant Healthcare Inc., Eden Prairie, MN, USA) were used for surface warming, the IV fluids were warmed, and surgical wound closure and dressing were performed rapidly to prevent surface heat loss. The first arterial blood gas analysis after ventricular fibrillation, calculated as for a body temperature of 37°C, showed the following: pH, 7.690; PCO2, 15 mmHg; PO2, 260 mmHg; HCO3, 18.1 mmol/L; base excess, −1.2; lactate, 0.33 mmol/L; ionized calcium, 0.88 mmol/L; and potassium, 5.0 mmol/L. Follow-up arterial blood gas analyses were repeated periodically. As we continued to perform CPR, the patient developed severe metabolic acidosis and hypocalcemia characterized by a pH of 6.840, HCO3 of 15.4 mmol/L, and ionized calcium of 0.66 mmol/L. To correct the abnormal findings, we administered appropriate amounts of sodium bicarbonate and calcium chloride. After an hour-long CPR and rewarming, the patient’s esophageal temperature recovered to 33.0°C, which was followed by observing sinus rhythm characterized by an HR of 75 bpm on an ECG monitor and detecting a pulse in the femoral artery with a characteristic ABP waveform that eventually led to the recovery of BP to 130/55 mmHg. Upon confirming recovery of spontaneous circulation (ROSC), the patient was transported to the ICU while we administered dopamine (10 µg/kg/minute), norepinephrine (0.1 µg/kg/minute), and vasopressin (2 U/hour). ABP, HR, and esophageal temperature during anesthesia are shown in Figure 5.
Figure 5. Changes in vital signs (blood pressure (BP), heart rate (HR), and esophageal temperature) over time (5-minute intervals).
bpm, beats per minute.
Cardiac biomarker testing performed after the patient’s transfer to the ICU showed an increased troponin-T level of 1.432 µg/L. Portable Doppler transthoracic echocardiography (TTE) revealed total occlusion of the left main coronary artery, which had undergone PCI previously. Because emergency coronary artery bypass grafting (CABG) was impossible at the time, repeated PCI was considered an alternative treatment option. However, even after his transfer to the ICU, the patient experienced repetitive ventricular fibrillation. Four hours after the patient’s transfer to the ICU, the patient’s guardian requested that CPR should be stopped. Without resuscitation, the patient died.
Discussion
In this case, the patient, who was receiving therapeutic hypothermia, underwent emergency craniectomy owing to worsening brain hemorrhage, and cardiac arrest occurred during general anesthesia. We deduced that the cardiac arrest was owing to the additional decrease in the patient’s core body temperature and myocardial infarction caused by restenosis of the previous PCI-treated lesion.
Therapeutic hypothermia is used to improve neurological outcomes in cardiac arrest, TBI, stroke, and hypoxic–ischemic encephalopathy.1,4 Studies on the efficacy of therapeutic hypothermia for TBI have been less convincing, and they show conflicting results. However, hypothermia is consistently used as an option for TBI because it reduces intracranial pressure and the cerebral metabolic rate, thereby protecting neuronal function in the brain and decreasing the wide range of injuries or reparative processes by attenuating temperature-sensitive mechanisms, such as excitotoxicity, free radical generation, apoptosis, and inflammation.1,2
The cardiovascular effects associated with hypothermia are complex. Hypothermia affects the electrical conduction of myocardial cells, which can induce hypotension or bradycardia and even increase the risk of ventricular fibrillation by prolonging repolarization.4,8 However, mild hypothermia of 32.0°C to 35.0°C demonstrates cardioprotective effects by decreasing myocardial oxygen demand and increasing myocardial perfusion through vasodilation of coronary vessels.9 Therefore, there has been a growing interest in the clinical use of mild therapeutic hypothermia as an adjunctive therapy during ischemic heart disease.10 Our patient received mild therapeutic hypothermia with a target core temperature of 33.0°C, which may have yielded cardioprotective as well as neuroprotective benefits.11
During general anesthesia, hypothermia is induced following an increase in heat loss caused by peripheral vasodilation after the injection of anesthetic drugs, heat redistribution from the core to the periphery, and impaired normal thermoregulatory responses, such as vasoconstriction or shivering.5 In our patient, the cold operating room temperature and irrigating an uncovered surgical area in the brain with cold saline also may have contributed to the additional core body temperature drop in addition to the effects of general anesthesia.12 However, in our case, neurosurgeons requested that anesthesiologists not perform warming during anesthesia. Instead, esophageal temperature and signs that could result from hypothermia, such as ECG changes, were carefully monitored. The esophageal temperature decreased from 33.0°C, which was measured immediately after entering the operation room, to 32.0°C at the time of cardiac arrest. The esophageal temperature further decreased to a minimum of 30.0°C during CPR. With rewarming, the temperature recovered to 33.0°C and finally, ROSC was achieved. Considering that chest compressions, drugs, and defibrillation did not induce ROSC, which was achieved with the recovery of an esophageal temperature of 33.0°C, the additional rapid fall in the patient’s core body temperature during anesthesia may have caused the cardiac arrest.
In this case study, we suspected that restenosis at the site of the previous stent insertion contributed to the cardiac arrest, given the following three findings: First, the patient had a normal preoperative ECG with no signs of ischemia but developed ST-segment depression during anesthesia. Second, repetitive cardiac arrests occurred even after achieving ROSC with rewarming. Third, an increased troponin-T concentration was observed, and suspected re-stenotic occlusion of the left main coronary artery was confirmed by Doppler TTE after the patient’s transfer to the ICU. Accordingly, myocardial infarction was considered the main cause of death.
The mechanisms of myocardial infarction owing to stent restenosis after DES insertion are rapid stent thrombosis or profuse in-stent restenosis (ISR).13 We assume that the lesion characteristics in the left main bifurcation of the coronary artery and DAPT interruption owing to the increasing amount of cerebral hemorrhage contributed to the stent restenosis in this patient. As evidence in support of this assumption, previous studies have shown that stenting at the left main bifurcation and DAPT interruption are strong risk factors for stent thrombosis or ISR.14,15 While controversial, not interrupting aspirin perioperatively could have avoided poor outcomes.16 However, there were no clinical signs related to expected ongoing restenosis of the coronary artery or myocardial ischemia preoperatively in our patient. We suspect that the possible ischemic signs may have been masked by the aforementioned cardioprotective benefits of therapeutic hypothermia. We also believe that the following three conditions contributed to the intraoperative myocardial ischemia and infarction caused by acute restenosis of the coronary stent: First, intraoperative hypotension may have contributed to myocardial injury. The IV anesthetics used during anesthesia, propofol and remifentanil, induced redistribution of blood volume by peripheral vasodilation, resulting in profound hypotension.4 Profound hypotension was strongly associated with myocardial injury and especially harmful for this patient, who had preexisting ischemic heart disease.17 Second, tachycardia, which may have been caused by the inotrope dopamine, used to compensate for the low BP or the stress response to surgical stimulation from shallow anesthetic depth, may have triggered myocardial ischemia by decreasing oxygen delivery and increasing oxygen consumption.17,18 If cardiogenic shock is suspected, as in our case, norepinephrine could be a better option than dopamine.19 Third, the bolus injection of phenylephrine, a selective α1-adrenergic agonist, just before cardiac arrest may have temporarily and abruptly increased cardiac afterload and impaired myocardial perfusion.20
Had we known that coronary artery occlusion would result from hypothermia-induced cardiac arrest and coronary stent restenosis, the treatment of choice would have been CABG surgery while maintaining circulation with emergency cardiopulmonary bypass (CPB) or extracorporeal membrane oxygenation (ECMO) via the femoral artery and vein.11 However, it was difficult to determine the etiology of the cardiac arrest at the time. Even if we had known the cause, ideal treatment would have been impossible, given circumstances beyond our control; all thoracic surgeons were busy operating, and emergency CPB and ECMO were not available. As an alternative, repeat PCI after the patient achieved ROSC could have been performed. However, the patient had a prolonged CPR duration, and he did not maintain the ROSC state for a period sufficient for intervention, which led to a poor outcome.
In this case, it was not possible to accurately determine the reason for the cardiac arrest and for not achieving consistent ROSC. Instead, we conclude that cardiac arrest developed and persisted as a result of the unintentional severe hypothermia that occurred during general anesthesia and/or coronary artery restenosis at the site of the previous PCI. A thorough evaluation of heart-related history is required for TBI patients who have a history of ischemic heart diseases and who undergo therapeutic hypothermia because there may be undiscovered coronary insufficiency owing to the cardioprotective effects of therapeutic hypothermia. In addition, careful monitoring of the patient’s core body temperature, and preventive and countermeasures should be fully prepared and taken against adverse hemodynamic effects, even cardiac arrest, which can occur because of additive hypothermic responses during anesthesia.
Ethics statement: This case report was approved by our Institutional Review Board (EMC 2020-05-005-001), and the case report was performed in compliance with the EQUATOR Network guidelines (the CARE guidelines). All patient details were de-identified, and written consent was obtained from the patient’s guardian for treatment and publication of this report.
Declaration of conflicting interest: The authors declare that there is no conflict of interest.
Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
ORCID iDs: Dong Ho Park https://orcid.org/0000-0002-6587-3756
Tae Woo Kim https://orcid.org/0000-0002-2115-6448
Mo Se Kim https://orcid.org/0000-0001-7362-0030 | CALCIUM CHLORIDE, DOPAMINE HYDROCHLORIDE, EPINEPHRINE, ISOSORBIDE DINITRATE, NOREPINEPHRINE, PHENYLEPHRINE HYDROCHLORIDE, PROPOFOL, REMIFENTANIL, ROCURONIUM BROMIDE, SODIUM BICARBONATE, SODIUM CHLORIDE, VASOPRESSIN | DrugsGivenReaction | CC BY-NC | 33499678 | 19,395,621 | 2021-01 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Metabolic acidosis'. | Cardiac arrest caused by accidental severe hypothermia and myocardial infarction during general anesthesia.
Therapeutic hypothermia is often used for traumatic brain injury because of its neuroprotective effect and decreased secondary brain injury. However, this procedure lacks clinical evidence supporting its efficacy, and adverse outcomes have been reported during general anesthesia. A 61-year-old man with a history of percutaneous coronary intervention (PCI) was admitted with traumatic brain injury. Immediately after admission, he underwent mild therapeutic hypothermia with a target temperature of 33.0°C for neuroprotection. During general anesthesia for emergency surgery because he developed a mass effect, hypothermic cardiac arrest occurred following an additional decrease in the core body temperature. Moreover, myocardial infarction caused by restenosis of the previous PCI lesion also contributed to the cardiac arrest. Although the patient recovered spontaneous circulation after an hour-long cardiopulmonary resuscitation with rewarming, he eventually died of subsequent repetitive cardiac arrests. When anesthetizing patients undergoing therapeutic hypothermia, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.
Introduction
Therapeutic hypothermia is an attractive strategy for traumatic brain injury (TBI), which demonstrates neuroprotective effects and decreases secondary brain injury.1 Despite the lack of clinical evidence regarding efficacy, therapeutic hypothermia after TBI has been used for at least half a century because of its potential benefits, which are supported by numerous preclinical studies.2,3 However, there have been no reports on the danger of unintentional severe hypothermia during general anesthesia caused by the additional drop in temperature of patients undergoing therapeutic hypothermia. When anesthetizing patients with comorbidities, such as ischemic heart disease, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.4,5
Case report
A 61-year-old man (height, 168 cm; weight, 70 kg) presented to our emergency room (ER) with head trauma after falling on a cement floor from a 5-meter-tall construction site. Upon arrival at the ER, the patient had a Glasgow coma scale score of 11 and the following vital signs: blood pressure (BP), 147/83 mmHg; heart rate (HR), 86 beats per minute (bpm); respiratory rate, 24 breaths per minute; body temperature from the auditory canal, 36.8°C; and percutaneous oxygen saturation (SpO2), 97%. Upon discovering signs of acute subdural hemorrhage, subarachnoid hemorrhage, and intraventricular hemorrhage from a brain computed tomography (CT) scan performed in the ER, we admitted the patient to the intensive care unit (ICU) for close monitoring. After approximately 6 hours, we performed follow-up brain CT because the patient’s mental status had deteriorated from drowsy to stupor. Upon observing a mass effect characterized by an increase in the hemorrhage volume and a left midline shift on the CT images, the patient was scheduled for immediate intubation and emergency decompressive craniectomy (Figure 1).
Figure 1. Brain computed tomography (CT) images taken 6 hours after intensive care unit admission. (a) Axial view showing acute subarachnoid hemorrhage in the peri-mesencephalic cistern (arrow A) and subdural hemorrhage in the right inferior temporal (arrow B) and right parietal (arrow C) regions. (b) Axial view showing acute subdural hemorrhage in the right temporal region (black arrow) and left midline shift.
Although the patient had a history of ischemic heart disease with percutaneous coronary intervention (PCI), it was difficult to obtain accurate information because there was no relevant data in the patient’s medical records from the authors’ hospital, and his drowsy state interfered with detailed interviews. According to the chest CT scan performed immediately after admission, coronary stents had been inserted at the bifurcation site of the left main coronary artery (Figure 2). It also appeared that the inserted stents were drug-eluting stents (DES) considering that he had received dual antiplatelet therapy (DAPT) with aspirin and clopidogrel (Plavix®; Sanofi, Bridgewater, NJ, USA) a switching platelet P2Y12 inhibitor, for approximately 4 months. However, DAPT was discontinued owing to the worsening cerebral hemorrhage after hospitalization; bridging antiplatelet therapy was also not performed. Before the surgery, the patient showed a normal electrocardiogram (ECG), and cardiac biomarker testing revealed a troponin-T concentration of 0.006 µg/L, which was within normal limits (Figure 3). The patient demonstrated satisfactory functional capacity to be able to work at a construction site just before the trauma. Therefore, an additional preoperative cardiac evaluation was not considered necessary.
Figure 2. Chest computed tomography (CT) images showing coronary stents in the main bifurcation of the left coronary arteries (white arrow in (a) and (b)).
Figure 3. Preoperative electrocardiogram of the patient showing normal sinus rhythm with a heart rate of 95 bpm.
bpm, beats per minute.
Immediately after admission to the ICU, the patient’s bladder temperature was 38.0°C. Using a hydrogel-coated water-circulating pad (Arctic Sun®; Medivance Inc., Louisville, CO, USA), which is a surface cooling device, the patient underwent mild therapeutic hypothermia with a target temperature of 33.0°C. During therapeutic hypothermia, sedation and shivering control were performed using continuous rate infusions of midazolam at 5 mg/hour and remifentanil at 0.2 µg/kg/minute. The patient’s bladder temperature was 33.4°C just before he was transferred to the operating room. Preoperative laboratory tests showed no unusual findings except for mildly elevated aspartate aminotransferase (73 U/L) and alanine aminotransferase (51 U/L) concentrations.
The patient’s vital signs upon arrival at the operating room were a BP of 130/70 mmHg and SpO2 of 100%. Lead II of the ECG was monitored, and there were no unusual findings except sinus tachycardia with an HR of 100 bpm on the initial ECG. Arterial BP (ABP), central venous pressure (CVP), and esophageal temperature were monitored from the right radial artery, right subclavian vein, and lower third of the esophagus using a temperature sensor attached to the esophageal stethoscope, respectively. The initial ABP, CVP, and esophageal temperature were 140/85 mmHg, 6 mmHg, and 33.0°C, respectively. Arterial blood gas analysis was performed in the arterial line, with values calculated as for a body temperature of 37°C, which revealed the following: pH, 7.400; partial pressure of carbon dioxide (PCO2), 40 mmHg; partial pressure of oxygen (PO2), 174 mmHg; bicarbonate (HCO3), 24.8 mmol/L; base excess, 0.0; lactate, 0.17 mmol/L; ionized calcium, 1.04 mmol/L; and potassium, 3.3 mmol/L. Total intravenous (IV) anesthesia was performed using a target-controlled infusion of propofol and remifentanil in accordance with the Schnider6 and Minto7 models, respectively. Propofol and remifentanil were infused continuously with effect-site concentrations (Ce) of 1.5 µg/mL and 1.5 ng/mL, respectively. Concurrently, 50 mg of rocuronium was administered by IV bolus injection for neuromuscular blockade. Mechanical ventilation was adjusted to maintain the end-tidal carbon dioxide value at 30 to 33 mmHg.
As ABP decreased to 78/42 mmHg and CVP decreased to 4 mmHg within 10 minutes after anesthesia, the Ce of propofol and remifentanil were lowered to 0.8 µg/mL and 1.0 ng/mL, respectively, and approximately 1 L of balanced crystalloid fluid loading was performed. However, the patient’s BP did not recover, and a dopamine continuous infusion (5 µg/kg/minute) was initiated. When a surgical incision was made 10 minutes after beginning the dopamine infusion, the patient’s ABP and HR increased to 115/70 mmHg and 120 bpm, respectively, and marked ST-segment depression (−2.4 mm) was found, which was not observed on the preoperative or initial ECGs (Figure 4). Subsequently, the depth of anesthesia was increased by increasing the Ce of propofol to 1.0 to 1.2 µg/mL, followed by a continuous infusion of isosorbide dinitrate (1 µg/kg/minute). While HR was maintained at 110 to 120 bpm, ABP gradually decreased to 70/40 mmHg after approximately 5 minutes. To treat the hypotension, isosorbide dinitrate was discontinued, and a bolus injection of 100 µg phenylephrine was administered, after which, ABP increased temporarily to 95/60 mmHg. Subsequently, HR decreased suddenly and rapidly, and ventricular fibrillation occurred. The anesthesiologists immediately discontinued the surgery and performed cardiopulmonary resuscitation (CPR). The patient’s esophageal temperature at the time was 32.0°C.
Figure 4. Intraoperative electrocardiogram of the patient showing sinus tachycardia with significant ST depression. The ST depression on the monitor represented approximately −2.4 at the time.
Chest compressions were performed with periodic defibrillation with 200 J, and epinephrine injections were administered every 3 minutes; these measures elicited no response. During CPR, ECG monitoring showed temporary severe bradycardia (HR, 10–30 bpm). However, this was considered pulseless electrical activity because no pulse was detected in the patient’s femoral artery. Soon afterward, there was a rapid transition to ventricular fibrillation, and CPR was repeated. At the time, the patient’s esophageal temperature dropped to 30.0°C. According to the refractory state after the defibrillation and drug injections, we considered that the cardiac arrest was caused by hypothermia. To reverse the cause, a forced-air warming device and a warming blanket (Bair Hugger®; Arizant Healthcare Inc., Eden Prairie, MN, USA) were used for surface warming, the IV fluids were warmed, and surgical wound closure and dressing were performed rapidly to prevent surface heat loss. The first arterial blood gas analysis after ventricular fibrillation, calculated as for a body temperature of 37°C, showed the following: pH, 7.690; PCO2, 15 mmHg; PO2, 260 mmHg; HCO3, 18.1 mmol/L; base excess, −1.2; lactate, 0.33 mmol/L; ionized calcium, 0.88 mmol/L; and potassium, 5.0 mmol/L. Follow-up arterial blood gas analyses were repeated periodically. As we continued to perform CPR, the patient developed severe metabolic acidosis and hypocalcemia characterized by a pH of 6.840, HCO3 of 15.4 mmol/L, and ionized calcium of 0.66 mmol/L. To correct the abnormal findings, we administered appropriate amounts of sodium bicarbonate and calcium chloride. After an hour-long CPR and rewarming, the patient’s esophageal temperature recovered to 33.0°C, which was followed by observing sinus rhythm characterized by an HR of 75 bpm on an ECG monitor and detecting a pulse in the femoral artery with a characteristic ABP waveform that eventually led to the recovery of BP to 130/55 mmHg. Upon confirming recovery of spontaneous circulation (ROSC), the patient was transported to the ICU while we administered dopamine (10 µg/kg/minute), norepinephrine (0.1 µg/kg/minute), and vasopressin (2 U/hour). ABP, HR, and esophageal temperature during anesthesia are shown in Figure 5.
Figure 5. Changes in vital signs (blood pressure (BP), heart rate (HR), and esophageal temperature) over time (5-minute intervals).
bpm, beats per minute.
Cardiac biomarker testing performed after the patient’s transfer to the ICU showed an increased troponin-T level of 1.432 µg/L. Portable Doppler transthoracic echocardiography (TTE) revealed total occlusion of the left main coronary artery, which had undergone PCI previously. Because emergency coronary artery bypass grafting (CABG) was impossible at the time, repeated PCI was considered an alternative treatment option. However, even after his transfer to the ICU, the patient experienced repetitive ventricular fibrillation. Four hours after the patient’s transfer to the ICU, the patient’s guardian requested that CPR should be stopped. Without resuscitation, the patient died.
Discussion
In this case, the patient, who was receiving therapeutic hypothermia, underwent emergency craniectomy owing to worsening brain hemorrhage, and cardiac arrest occurred during general anesthesia. We deduced that the cardiac arrest was owing to the additional decrease in the patient’s core body temperature and myocardial infarction caused by restenosis of the previous PCI-treated lesion.
Therapeutic hypothermia is used to improve neurological outcomes in cardiac arrest, TBI, stroke, and hypoxic–ischemic encephalopathy.1,4 Studies on the efficacy of therapeutic hypothermia for TBI have been less convincing, and they show conflicting results. However, hypothermia is consistently used as an option for TBI because it reduces intracranial pressure and the cerebral metabolic rate, thereby protecting neuronal function in the brain and decreasing the wide range of injuries or reparative processes by attenuating temperature-sensitive mechanisms, such as excitotoxicity, free radical generation, apoptosis, and inflammation.1,2
The cardiovascular effects associated with hypothermia are complex. Hypothermia affects the electrical conduction of myocardial cells, which can induce hypotension or bradycardia and even increase the risk of ventricular fibrillation by prolonging repolarization.4,8 However, mild hypothermia of 32.0°C to 35.0°C demonstrates cardioprotective effects by decreasing myocardial oxygen demand and increasing myocardial perfusion through vasodilation of coronary vessels.9 Therefore, there has been a growing interest in the clinical use of mild therapeutic hypothermia as an adjunctive therapy during ischemic heart disease.10 Our patient received mild therapeutic hypothermia with a target core temperature of 33.0°C, which may have yielded cardioprotective as well as neuroprotective benefits.11
During general anesthesia, hypothermia is induced following an increase in heat loss caused by peripheral vasodilation after the injection of anesthetic drugs, heat redistribution from the core to the periphery, and impaired normal thermoregulatory responses, such as vasoconstriction or shivering.5 In our patient, the cold operating room temperature and irrigating an uncovered surgical area in the brain with cold saline also may have contributed to the additional core body temperature drop in addition to the effects of general anesthesia.12 However, in our case, neurosurgeons requested that anesthesiologists not perform warming during anesthesia. Instead, esophageal temperature and signs that could result from hypothermia, such as ECG changes, were carefully monitored. The esophageal temperature decreased from 33.0°C, which was measured immediately after entering the operation room, to 32.0°C at the time of cardiac arrest. The esophageal temperature further decreased to a minimum of 30.0°C during CPR. With rewarming, the temperature recovered to 33.0°C and finally, ROSC was achieved. Considering that chest compressions, drugs, and defibrillation did not induce ROSC, which was achieved with the recovery of an esophageal temperature of 33.0°C, the additional rapid fall in the patient’s core body temperature during anesthesia may have caused the cardiac arrest.
In this case study, we suspected that restenosis at the site of the previous stent insertion contributed to the cardiac arrest, given the following three findings: First, the patient had a normal preoperative ECG with no signs of ischemia but developed ST-segment depression during anesthesia. Second, repetitive cardiac arrests occurred even after achieving ROSC with rewarming. Third, an increased troponin-T concentration was observed, and suspected re-stenotic occlusion of the left main coronary artery was confirmed by Doppler TTE after the patient’s transfer to the ICU. Accordingly, myocardial infarction was considered the main cause of death.
The mechanisms of myocardial infarction owing to stent restenosis after DES insertion are rapid stent thrombosis or profuse in-stent restenosis (ISR).13 We assume that the lesion characteristics in the left main bifurcation of the coronary artery and DAPT interruption owing to the increasing amount of cerebral hemorrhage contributed to the stent restenosis in this patient. As evidence in support of this assumption, previous studies have shown that stenting at the left main bifurcation and DAPT interruption are strong risk factors for stent thrombosis or ISR.14,15 While controversial, not interrupting aspirin perioperatively could have avoided poor outcomes.16 However, there were no clinical signs related to expected ongoing restenosis of the coronary artery or myocardial ischemia preoperatively in our patient. We suspect that the possible ischemic signs may have been masked by the aforementioned cardioprotective benefits of therapeutic hypothermia. We also believe that the following three conditions contributed to the intraoperative myocardial ischemia and infarction caused by acute restenosis of the coronary stent: First, intraoperative hypotension may have contributed to myocardial injury. The IV anesthetics used during anesthesia, propofol and remifentanil, induced redistribution of blood volume by peripheral vasodilation, resulting in profound hypotension.4 Profound hypotension was strongly associated with myocardial injury and especially harmful for this patient, who had preexisting ischemic heart disease.17 Second, tachycardia, which may have been caused by the inotrope dopamine, used to compensate for the low BP or the stress response to surgical stimulation from shallow anesthetic depth, may have triggered myocardial ischemia by decreasing oxygen delivery and increasing oxygen consumption.17,18 If cardiogenic shock is suspected, as in our case, norepinephrine could be a better option than dopamine.19 Third, the bolus injection of phenylephrine, a selective α1-adrenergic agonist, just before cardiac arrest may have temporarily and abruptly increased cardiac afterload and impaired myocardial perfusion.20
Had we known that coronary artery occlusion would result from hypothermia-induced cardiac arrest and coronary stent restenosis, the treatment of choice would have been CABG surgery while maintaining circulation with emergency cardiopulmonary bypass (CPB) or extracorporeal membrane oxygenation (ECMO) via the femoral artery and vein.11 However, it was difficult to determine the etiology of the cardiac arrest at the time. Even if we had known the cause, ideal treatment would have been impossible, given circumstances beyond our control; all thoracic surgeons were busy operating, and emergency CPB and ECMO were not available. As an alternative, repeat PCI after the patient achieved ROSC could have been performed. However, the patient had a prolonged CPR duration, and he did not maintain the ROSC state for a period sufficient for intervention, which led to a poor outcome.
In this case, it was not possible to accurately determine the reason for the cardiac arrest and for not achieving consistent ROSC. Instead, we conclude that cardiac arrest developed and persisted as a result of the unintentional severe hypothermia that occurred during general anesthesia and/or coronary artery restenosis at the site of the previous PCI. A thorough evaluation of heart-related history is required for TBI patients who have a history of ischemic heart diseases and who undergo therapeutic hypothermia because there may be undiscovered coronary insufficiency owing to the cardioprotective effects of therapeutic hypothermia. In addition, careful monitoring of the patient’s core body temperature, and preventive and countermeasures should be fully prepared and taken against adverse hemodynamic effects, even cardiac arrest, which can occur because of additive hypothermic responses during anesthesia.
Ethics statement: This case report was approved by our Institutional Review Board (EMC 2020-05-005-001), and the case report was performed in compliance with the EQUATOR Network guidelines (the CARE guidelines). All patient details were de-identified, and written consent was obtained from the patient’s guardian for treatment and publication of this report.
Declaration of conflicting interest: The authors declare that there is no conflict of interest.
Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
ORCID iDs: Dong Ho Park https://orcid.org/0000-0002-6587-3756
Tae Woo Kim https://orcid.org/0000-0002-2115-6448
Mo Se Kim https://orcid.org/0000-0001-7362-0030 | DOPAMINE HYDROCHLORIDE, MIDAZOLAM, PHENYLEPHRINE HYDROCHLORIDE, PROPOFOL, REMIFENTANIL, REMIFENTANIL HYDROCHLORIDE, ROCURONIUM BROMIDE, SODIUM CHLORIDE | DrugsGivenReaction | CC BY-NC | 33499678 | 19,429,456 | 2021-01 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Myocardial ischaemia'. | Cardiac arrest caused by accidental severe hypothermia and myocardial infarction during general anesthesia.
Therapeutic hypothermia is often used for traumatic brain injury because of its neuroprotective effect and decreased secondary brain injury. However, this procedure lacks clinical evidence supporting its efficacy, and adverse outcomes have been reported during general anesthesia. A 61-year-old man with a history of percutaneous coronary intervention (PCI) was admitted with traumatic brain injury. Immediately after admission, he underwent mild therapeutic hypothermia with a target temperature of 33.0°C for neuroprotection. During general anesthesia for emergency surgery because he developed a mass effect, hypothermic cardiac arrest occurred following an additional decrease in the core body temperature. Moreover, myocardial infarction caused by restenosis of the previous PCI lesion also contributed to the cardiac arrest. Although the patient recovered spontaneous circulation after an hour-long cardiopulmonary resuscitation with rewarming, he eventually died of subsequent repetitive cardiac arrests. When anesthetizing patients undergoing therapeutic hypothermia, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.
Introduction
Therapeutic hypothermia is an attractive strategy for traumatic brain injury (TBI), which demonstrates neuroprotective effects and decreases secondary brain injury.1 Despite the lack of clinical evidence regarding efficacy, therapeutic hypothermia after TBI has been used for at least half a century because of its potential benefits, which are supported by numerous preclinical studies.2,3 However, there have been no reports on the danger of unintentional severe hypothermia during general anesthesia caused by the additional drop in temperature of patients undergoing therapeutic hypothermia. When anesthetizing patients with comorbidities, such as ischemic heart disease, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.4,5
Case report
A 61-year-old man (height, 168 cm; weight, 70 kg) presented to our emergency room (ER) with head trauma after falling on a cement floor from a 5-meter-tall construction site. Upon arrival at the ER, the patient had a Glasgow coma scale score of 11 and the following vital signs: blood pressure (BP), 147/83 mmHg; heart rate (HR), 86 beats per minute (bpm); respiratory rate, 24 breaths per minute; body temperature from the auditory canal, 36.8°C; and percutaneous oxygen saturation (SpO2), 97%. Upon discovering signs of acute subdural hemorrhage, subarachnoid hemorrhage, and intraventricular hemorrhage from a brain computed tomography (CT) scan performed in the ER, we admitted the patient to the intensive care unit (ICU) for close monitoring. After approximately 6 hours, we performed follow-up brain CT because the patient’s mental status had deteriorated from drowsy to stupor. Upon observing a mass effect characterized by an increase in the hemorrhage volume and a left midline shift on the CT images, the patient was scheduled for immediate intubation and emergency decompressive craniectomy (Figure 1).
Figure 1. Brain computed tomography (CT) images taken 6 hours after intensive care unit admission. (a) Axial view showing acute subarachnoid hemorrhage in the peri-mesencephalic cistern (arrow A) and subdural hemorrhage in the right inferior temporal (arrow B) and right parietal (arrow C) regions. (b) Axial view showing acute subdural hemorrhage in the right temporal region (black arrow) and left midline shift.
Although the patient had a history of ischemic heart disease with percutaneous coronary intervention (PCI), it was difficult to obtain accurate information because there was no relevant data in the patient’s medical records from the authors’ hospital, and his drowsy state interfered with detailed interviews. According to the chest CT scan performed immediately after admission, coronary stents had been inserted at the bifurcation site of the left main coronary artery (Figure 2). It also appeared that the inserted stents were drug-eluting stents (DES) considering that he had received dual antiplatelet therapy (DAPT) with aspirin and clopidogrel (Plavix®; Sanofi, Bridgewater, NJ, USA) a switching platelet P2Y12 inhibitor, for approximately 4 months. However, DAPT was discontinued owing to the worsening cerebral hemorrhage after hospitalization; bridging antiplatelet therapy was also not performed. Before the surgery, the patient showed a normal electrocardiogram (ECG), and cardiac biomarker testing revealed a troponin-T concentration of 0.006 µg/L, which was within normal limits (Figure 3). The patient demonstrated satisfactory functional capacity to be able to work at a construction site just before the trauma. Therefore, an additional preoperative cardiac evaluation was not considered necessary.
Figure 2. Chest computed tomography (CT) images showing coronary stents in the main bifurcation of the left coronary arteries (white arrow in (a) and (b)).
Figure 3. Preoperative electrocardiogram of the patient showing normal sinus rhythm with a heart rate of 95 bpm.
bpm, beats per minute.
Immediately after admission to the ICU, the patient’s bladder temperature was 38.0°C. Using a hydrogel-coated water-circulating pad (Arctic Sun®; Medivance Inc., Louisville, CO, USA), which is a surface cooling device, the patient underwent mild therapeutic hypothermia with a target temperature of 33.0°C. During therapeutic hypothermia, sedation and shivering control were performed using continuous rate infusions of midazolam at 5 mg/hour and remifentanil at 0.2 µg/kg/minute. The patient’s bladder temperature was 33.4°C just before he was transferred to the operating room. Preoperative laboratory tests showed no unusual findings except for mildly elevated aspartate aminotransferase (73 U/L) and alanine aminotransferase (51 U/L) concentrations.
The patient’s vital signs upon arrival at the operating room were a BP of 130/70 mmHg and SpO2 of 100%. Lead II of the ECG was monitored, and there were no unusual findings except sinus tachycardia with an HR of 100 bpm on the initial ECG. Arterial BP (ABP), central venous pressure (CVP), and esophageal temperature were monitored from the right radial artery, right subclavian vein, and lower third of the esophagus using a temperature sensor attached to the esophageal stethoscope, respectively. The initial ABP, CVP, and esophageal temperature were 140/85 mmHg, 6 mmHg, and 33.0°C, respectively. Arterial blood gas analysis was performed in the arterial line, with values calculated as for a body temperature of 37°C, which revealed the following: pH, 7.400; partial pressure of carbon dioxide (PCO2), 40 mmHg; partial pressure of oxygen (PO2), 174 mmHg; bicarbonate (HCO3), 24.8 mmol/L; base excess, 0.0; lactate, 0.17 mmol/L; ionized calcium, 1.04 mmol/L; and potassium, 3.3 mmol/L. Total intravenous (IV) anesthesia was performed using a target-controlled infusion of propofol and remifentanil in accordance with the Schnider6 and Minto7 models, respectively. Propofol and remifentanil were infused continuously with effect-site concentrations (Ce) of 1.5 µg/mL and 1.5 ng/mL, respectively. Concurrently, 50 mg of rocuronium was administered by IV bolus injection for neuromuscular blockade. Mechanical ventilation was adjusted to maintain the end-tidal carbon dioxide value at 30 to 33 mmHg.
As ABP decreased to 78/42 mmHg and CVP decreased to 4 mmHg within 10 minutes after anesthesia, the Ce of propofol and remifentanil were lowered to 0.8 µg/mL and 1.0 ng/mL, respectively, and approximately 1 L of balanced crystalloid fluid loading was performed. However, the patient’s BP did not recover, and a dopamine continuous infusion (5 µg/kg/minute) was initiated. When a surgical incision was made 10 minutes after beginning the dopamine infusion, the patient’s ABP and HR increased to 115/70 mmHg and 120 bpm, respectively, and marked ST-segment depression (−2.4 mm) was found, which was not observed on the preoperative or initial ECGs (Figure 4). Subsequently, the depth of anesthesia was increased by increasing the Ce of propofol to 1.0 to 1.2 µg/mL, followed by a continuous infusion of isosorbide dinitrate (1 µg/kg/minute). While HR was maintained at 110 to 120 bpm, ABP gradually decreased to 70/40 mmHg after approximately 5 minutes. To treat the hypotension, isosorbide dinitrate was discontinued, and a bolus injection of 100 µg phenylephrine was administered, after which, ABP increased temporarily to 95/60 mmHg. Subsequently, HR decreased suddenly and rapidly, and ventricular fibrillation occurred. The anesthesiologists immediately discontinued the surgery and performed cardiopulmonary resuscitation (CPR). The patient’s esophageal temperature at the time was 32.0°C.
Figure 4. Intraoperative electrocardiogram of the patient showing sinus tachycardia with significant ST depression. The ST depression on the monitor represented approximately −2.4 at the time.
Chest compressions were performed with periodic defibrillation with 200 J, and epinephrine injections were administered every 3 minutes; these measures elicited no response. During CPR, ECG monitoring showed temporary severe bradycardia (HR, 10–30 bpm). However, this was considered pulseless electrical activity because no pulse was detected in the patient’s femoral artery. Soon afterward, there was a rapid transition to ventricular fibrillation, and CPR was repeated. At the time, the patient’s esophageal temperature dropped to 30.0°C. According to the refractory state after the defibrillation and drug injections, we considered that the cardiac arrest was caused by hypothermia. To reverse the cause, a forced-air warming device and a warming blanket (Bair Hugger®; Arizant Healthcare Inc., Eden Prairie, MN, USA) were used for surface warming, the IV fluids were warmed, and surgical wound closure and dressing were performed rapidly to prevent surface heat loss. The first arterial blood gas analysis after ventricular fibrillation, calculated as for a body temperature of 37°C, showed the following: pH, 7.690; PCO2, 15 mmHg; PO2, 260 mmHg; HCO3, 18.1 mmol/L; base excess, −1.2; lactate, 0.33 mmol/L; ionized calcium, 0.88 mmol/L; and potassium, 5.0 mmol/L. Follow-up arterial blood gas analyses were repeated periodically. As we continued to perform CPR, the patient developed severe metabolic acidosis and hypocalcemia characterized by a pH of 6.840, HCO3 of 15.4 mmol/L, and ionized calcium of 0.66 mmol/L. To correct the abnormal findings, we administered appropriate amounts of sodium bicarbonate and calcium chloride. After an hour-long CPR and rewarming, the patient’s esophageal temperature recovered to 33.0°C, which was followed by observing sinus rhythm characterized by an HR of 75 bpm on an ECG monitor and detecting a pulse in the femoral artery with a characteristic ABP waveform that eventually led to the recovery of BP to 130/55 mmHg. Upon confirming recovery of spontaneous circulation (ROSC), the patient was transported to the ICU while we administered dopamine (10 µg/kg/minute), norepinephrine (0.1 µg/kg/minute), and vasopressin (2 U/hour). ABP, HR, and esophageal temperature during anesthesia are shown in Figure 5.
Figure 5. Changes in vital signs (blood pressure (BP), heart rate (HR), and esophageal temperature) over time (5-minute intervals).
bpm, beats per minute.
Cardiac biomarker testing performed after the patient’s transfer to the ICU showed an increased troponin-T level of 1.432 µg/L. Portable Doppler transthoracic echocardiography (TTE) revealed total occlusion of the left main coronary artery, which had undergone PCI previously. Because emergency coronary artery bypass grafting (CABG) was impossible at the time, repeated PCI was considered an alternative treatment option. However, even after his transfer to the ICU, the patient experienced repetitive ventricular fibrillation. Four hours after the patient’s transfer to the ICU, the patient’s guardian requested that CPR should be stopped. Without resuscitation, the patient died.
Discussion
In this case, the patient, who was receiving therapeutic hypothermia, underwent emergency craniectomy owing to worsening brain hemorrhage, and cardiac arrest occurred during general anesthesia. We deduced that the cardiac arrest was owing to the additional decrease in the patient’s core body temperature and myocardial infarction caused by restenosis of the previous PCI-treated lesion.
Therapeutic hypothermia is used to improve neurological outcomes in cardiac arrest, TBI, stroke, and hypoxic–ischemic encephalopathy.1,4 Studies on the efficacy of therapeutic hypothermia for TBI have been less convincing, and they show conflicting results. However, hypothermia is consistently used as an option for TBI because it reduces intracranial pressure and the cerebral metabolic rate, thereby protecting neuronal function in the brain and decreasing the wide range of injuries or reparative processes by attenuating temperature-sensitive mechanisms, such as excitotoxicity, free radical generation, apoptosis, and inflammation.1,2
The cardiovascular effects associated with hypothermia are complex. Hypothermia affects the electrical conduction of myocardial cells, which can induce hypotension or bradycardia and even increase the risk of ventricular fibrillation by prolonging repolarization.4,8 However, mild hypothermia of 32.0°C to 35.0°C demonstrates cardioprotective effects by decreasing myocardial oxygen demand and increasing myocardial perfusion through vasodilation of coronary vessels.9 Therefore, there has been a growing interest in the clinical use of mild therapeutic hypothermia as an adjunctive therapy during ischemic heart disease.10 Our patient received mild therapeutic hypothermia with a target core temperature of 33.0°C, which may have yielded cardioprotective as well as neuroprotective benefits.11
During general anesthesia, hypothermia is induced following an increase in heat loss caused by peripheral vasodilation after the injection of anesthetic drugs, heat redistribution from the core to the periphery, and impaired normal thermoregulatory responses, such as vasoconstriction or shivering.5 In our patient, the cold operating room temperature and irrigating an uncovered surgical area in the brain with cold saline also may have contributed to the additional core body temperature drop in addition to the effects of general anesthesia.12 However, in our case, neurosurgeons requested that anesthesiologists not perform warming during anesthesia. Instead, esophageal temperature and signs that could result from hypothermia, such as ECG changes, were carefully monitored. The esophageal temperature decreased from 33.0°C, which was measured immediately after entering the operation room, to 32.0°C at the time of cardiac arrest. The esophageal temperature further decreased to a minimum of 30.0°C during CPR. With rewarming, the temperature recovered to 33.0°C and finally, ROSC was achieved. Considering that chest compressions, drugs, and defibrillation did not induce ROSC, which was achieved with the recovery of an esophageal temperature of 33.0°C, the additional rapid fall in the patient’s core body temperature during anesthesia may have caused the cardiac arrest.
In this case study, we suspected that restenosis at the site of the previous stent insertion contributed to the cardiac arrest, given the following three findings: First, the patient had a normal preoperative ECG with no signs of ischemia but developed ST-segment depression during anesthesia. Second, repetitive cardiac arrests occurred even after achieving ROSC with rewarming. Third, an increased troponin-T concentration was observed, and suspected re-stenotic occlusion of the left main coronary artery was confirmed by Doppler TTE after the patient’s transfer to the ICU. Accordingly, myocardial infarction was considered the main cause of death.
The mechanisms of myocardial infarction owing to stent restenosis after DES insertion are rapid stent thrombosis or profuse in-stent restenosis (ISR).13 We assume that the lesion characteristics in the left main bifurcation of the coronary artery and DAPT interruption owing to the increasing amount of cerebral hemorrhage contributed to the stent restenosis in this patient. As evidence in support of this assumption, previous studies have shown that stenting at the left main bifurcation and DAPT interruption are strong risk factors for stent thrombosis or ISR.14,15 While controversial, not interrupting aspirin perioperatively could have avoided poor outcomes.16 However, there were no clinical signs related to expected ongoing restenosis of the coronary artery or myocardial ischemia preoperatively in our patient. We suspect that the possible ischemic signs may have been masked by the aforementioned cardioprotective benefits of therapeutic hypothermia. We also believe that the following three conditions contributed to the intraoperative myocardial ischemia and infarction caused by acute restenosis of the coronary stent: First, intraoperative hypotension may have contributed to myocardial injury. The IV anesthetics used during anesthesia, propofol and remifentanil, induced redistribution of blood volume by peripheral vasodilation, resulting in profound hypotension.4 Profound hypotension was strongly associated with myocardial injury and especially harmful for this patient, who had preexisting ischemic heart disease.17 Second, tachycardia, which may have been caused by the inotrope dopamine, used to compensate for the low BP or the stress response to surgical stimulation from shallow anesthetic depth, may have triggered myocardial ischemia by decreasing oxygen delivery and increasing oxygen consumption.17,18 If cardiogenic shock is suspected, as in our case, norepinephrine could be a better option than dopamine.19 Third, the bolus injection of phenylephrine, a selective α1-adrenergic agonist, just before cardiac arrest may have temporarily and abruptly increased cardiac afterload and impaired myocardial perfusion.20
Had we known that coronary artery occlusion would result from hypothermia-induced cardiac arrest and coronary stent restenosis, the treatment of choice would have been CABG surgery while maintaining circulation with emergency cardiopulmonary bypass (CPB) or extracorporeal membrane oxygenation (ECMO) via the femoral artery and vein.11 However, it was difficult to determine the etiology of the cardiac arrest at the time. Even if we had known the cause, ideal treatment would have been impossible, given circumstances beyond our control; all thoracic surgeons were busy operating, and emergency CPB and ECMO were not available. As an alternative, repeat PCI after the patient achieved ROSC could have been performed. However, the patient had a prolonged CPR duration, and he did not maintain the ROSC state for a period sufficient for intervention, which led to a poor outcome.
In this case, it was not possible to accurately determine the reason for the cardiac arrest and for not achieving consistent ROSC. Instead, we conclude that cardiac arrest developed and persisted as a result of the unintentional severe hypothermia that occurred during general anesthesia and/or coronary artery restenosis at the site of the previous PCI. A thorough evaluation of heart-related history is required for TBI patients who have a history of ischemic heart diseases and who undergo therapeutic hypothermia because there may be undiscovered coronary insufficiency owing to the cardioprotective effects of therapeutic hypothermia. In addition, careful monitoring of the patient’s core body temperature, and preventive and countermeasures should be fully prepared and taken against adverse hemodynamic effects, even cardiac arrest, which can occur because of additive hypothermic responses during anesthesia.
Ethics statement: This case report was approved by our Institutional Review Board (EMC 2020-05-005-001), and the case report was performed in compliance with the EQUATOR Network guidelines (the CARE guidelines). All patient details were de-identified, and written consent was obtained from the patient’s guardian for treatment and publication of this report.
Declaration of conflicting interest: The authors declare that there is no conflict of interest.
Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
ORCID iDs: Dong Ho Park https://orcid.org/0000-0002-6587-3756
Tae Woo Kim https://orcid.org/0000-0002-2115-6448
Mo Se Kim https://orcid.org/0000-0001-7362-0030 | DOPAMINE HYDROCHLORIDE, MIDAZOLAM, PHENYLEPHRINE HYDROCHLORIDE, PROPOFOL, REMIFENTANIL, REMIFENTANIL HYDROCHLORIDE, ROCURONIUM BROMIDE, SODIUM CHLORIDE | DrugsGivenReaction | CC BY-NC | 33499678 | 19,429,456 | 2021-01 |
What is the weight of the patient? | Cardiac arrest caused by accidental severe hypothermia and myocardial infarction during general anesthesia.
Therapeutic hypothermia is often used for traumatic brain injury because of its neuroprotective effect and decreased secondary brain injury. However, this procedure lacks clinical evidence supporting its efficacy, and adverse outcomes have been reported during general anesthesia. A 61-year-old man with a history of percutaneous coronary intervention (PCI) was admitted with traumatic brain injury. Immediately after admission, he underwent mild therapeutic hypothermia with a target temperature of 33.0°C for neuroprotection. During general anesthesia for emergency surgery because he developed a mass effect, hypothermic cardiac arrest occurred following an additional decrease in the core body temperature. Moreover, myocardial infarction caused by restenosis of the previous PCI lesion also contributed to the cardiac arrest. Although the patient recovered spontaneous circulation after an hour-long cardiopulmonary resuscitation with rewarming, he eventually died of subsequent repetitive cardiac arrests. When anesthetizing patients undergoing therapeutic hypothermia, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.
Introduction
Therapeutic hypothermia is an attractive strategy for traumatic brain injury (TBI), which demonstrates neuroprotective effects and decreases secondary brain injury.1 Despite the lack of clinical evidence regarding efficacy, therapeutic hypothermia after TBI has been used for at least half a century because of its potential benefits, which are supported by numerous preclinical studies.2,3 However, there have been no reports on the danger of unintentional severe hypothermia during general anesthesia caused by the additional drop in temperature of patients undergoing therapeutic hypothermia. When anesthetizing patients with comorbidities, such as ischemic heart disease, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.4,5
Case report
A 61-year-old man (height, 168 cm; weight, 70 kg) presented to our emergency room (ER) with head trauma after falling on a cement floor from a 5-meter-tall construction site. Upon arrival at the ER, the patient had a Glasgow coma scale score of 11 and the following vital signs: blood pressure (BP), 147/83 mmHg; heart rate (HR), 86 beats per minute (bpm); respiratory rate, 24 breaths per minute; body temperature from the auditory canal, 36.8°C; and percutaneous oxygen saturation (SpO2), 97%. Upon discovering signs of acute subdural hemorrhage, subarachnoid hemorrhage, and intraventricular hemorrhage from a brain computed tomography (CT) scan performed in the ER, we admitted the patient to the intensive care unit (ICU) for close monitoring. After approximately 6 hours, we performed follow-up brain CT because the patient’s mental status had deteriorated from drowsy to stupor. Upon observing a mass effect characterized by an increase in the hemorrhage volume and a left midline shift on the CT images, the patient was scheduled for immediate intubation and emergency decompressive craniectomy (Figure 1).
Figure 1. Brain computed tomography (CT) images taken 6 hours after intensive care unit admission. (a) Axial view showing acute subarachnoid hemorrhage in the peri-mesencephalic cistern (arrow A) and subdural hemorrhage in the right inferior temporal (arrow B) and right parietal (arrow C) regions. (b) Axial view showing acute subdural hemorrhage in the right temporal region (black arrow) and left midline shift.
Although the patient had a history of ischemic heart disease with percutaneous coronary intervention (PCI), it was difficult to obtain accurate information because there was no relevant data in the patient’s medical records from the authors’ hospital, and his drowsy state interfered with detailed interviews. According to the chest CT scan performed immediately after admission, coronary stents had been inserted at the bifurcation site of the left main coronary artery (Figure 2). It also appeared that the inserted stents were drug-eluting stents (DES) considering that he had received dual antiplatelet therapy (DAPT) with aspirin and clopidogrel (Plavix®; Sanofi, Bridgewater, NJ, USA) a switching platelet P2Y12 inhibitor, for approximately 4 months. However, DAPT was discontinued owing to the worsening cerebral hemorrhage after hospitalization; bridging antiplatelet therapy was also not performed. Before the surgery, the patient showed a normal electrocardiogram (ECG), and cardiac biomarker testing revealed a troponin-T concentration of 0.006 µg/L, which was within normal limits (Figure 3). The patient demonstrated satisfactory functional capacity to be able to work at a construction site just before the trauma. Therefore, an additional preoperative cardiac evaluation was not considered necessary.
Figure 2. Chest computed tomography (CT) images showing coronary stents in the main bifurcation of the left coronary arteries (white arrow in (a) and (b)).
Figure 3. Preoperative electrocardiogram of the patient showing normal sinus rhythm with a heart rate of 95 bpm.
bpm, beats per minute.
Immediately after admission to the ICU, the patient’s bladder temperature was 38.0°C. Using a hydrogel-coated water-circulating pad (Arctic Sun®; Medivance Inc., Louisville, CO, USA), which is a surface cooling device, the patient underwent mild therapeutic hypothermia with a target temperature of 33.0°C. During therapeutic hypothermia, sedation and shivering control were performed using continuous rate infusions of midazolam at 5 mg/hour and remifentanil at 0.2 µg/kg/minute. The patient’s bladder temperature was 33.4°C just before he was transferred to the operating room. Preoperative laboratory tests showed no unusual findings except for mildly elevated aspartate aminotransferase (73 U/L) and alanine aminotransferase (51 U/L) concentrations.
The patient’s vital signs upon arrival at the operating room were a BP of 130/70 mmHg and SpO2 of 100%. Lead II of the ECG was monitored, and there were no unusual findings except sinus tachycardia with an HR of 100 bpm on the initial ECG. Arterial BP (ABP), central venous pressure (CVP), and esophageal temperature were monitored from the right radial artery, right subclavian vein, and lower third of the esophagus using a temperature sensor attached to the esophageal stethoscope, respectively. The initial ABP, CVP, and esophageal temperature were 140/85 mmHg, 6 mmHg, and 33.0°C, respectively. Arterial blood gas analysis was performed in the arterial line, with values calculated as for a body temperature of 37°C, which revealed the following: pH, 7.400; partial pressure of carbon dioxide (PCO2), 40 mmHg; partial pressure of oxygen (PO2), 174 mmHg; bicarbonate (HCO3), 24.8 mmol/L; base excess, 0.0; lactate, 0.17 mmol/L; ionized calcium, 1.04 mmol/L; and potassium, 3.3 mmol/L. Total intravenous (IV) anesthesia was performed using a target-controlled infusion of propofol and remifentanil in accordance with the Schnider6 and Minto7 models, respectively. Propofol and remifentanil were infused continuously with effect-site concentrations (Ce) of 1.5 µg/mL and 1.5 ng/mL, respectively. Concurrently, 50 mg of rocuronium was administered by IV bolus injection for neuromuscular blockade. Mechanical ventilation was adjusted to maintain the end-tidal carbon dioxide value at 30 to 33 mmHg.
As ABP decreased to 78/42 mmHg and CVP decreased to 4 mmHg within 10 minutes after anesthesia, the Ce of propofol and remifentanil were lowered to 0.8 µg/mL and 1.0 ng/mL, respectively, and approximately 1 L of balanced crystalloid fluid loading was performed. However, the patient’s BP did not recover, and a dopamine continuous infusion (5 µg/kg/minute) was initiated. When a surgical incision was made 10 minutes after beginning the dopamine infusion, the patient’s ABP and HR increased to 115/70 mmHg and 120 bpm, respectively, and marked ST-segment depression (−2.4 mm) was found, which was not observed on the preoperative or initial ECGs (Figure 4). Subsequently, the depth of anesthesia was increased by increasing the Ce of propofol to 1.0 to 1.2 µg/mL, followed by a continuous infusion of isosorbide dinitrate (1 µg/kg/minute). While HR was maintained at 110 to 120 bpm, ABP gradually decreased to 70/40 mmHg after approximately 5 minutes. To treat the hypotension, isosorbide dinitrate was discontinued, and a bolus injection of 100 µg phenylephrine was administered, after which, ABP increased temporarily to 95/60 mmHg. Subsequently, HR decreased suddenly and rapidly, and ventricular fibrillation occurred. The anesthesiologists immediately discontinued the surgery and performed cardiopulmonary resuscitation (CPR). The patient’s esophageal temperature at the time was 32.0°C.
Figure 4. Intraoperative electrocardiogram of the patient showing sinus tachycardia with significant ST depression. The ST depression on the monitor represented approximately −2.4 at the time.
Chest compressions were performed with periodic defibrillation with 200 J, and epinephrine injections were administered every 3 minutes; these measures elicited no response. During CPR, ECG monitoring showed temporary severe bradycardia (HR, 10–30 bpm). However, this was considered pulseless electrical activity because no pulse was detected in the patient’s femoral artery. Soon afterward, there was a rapid transition to ventricular fibrillation, and CPR was repeated. At the time, the patient’s esophageal temperature dropped to 30.0°C. According to the refractory state after the defibrillation and drug injections, we considered that the cardiac arrest was caused by hypothermia. To reverse the cause, a forced-air warming device and a warming blanket (Bair Hugger®; Arizant Healthcare Inc., Eden Prairie, MN, USA) were used for surface warming, the IV fluids were warmed, and surgical wound closure and dressing were performed rapidly to prevent surface heat loss. The first arterial blood gas analysis after ventricular fibrillation, calculated as for a body temperature of 37°C, showed the following: pH, 7.690; PCO2, 15 mmHg; PO2, 260 mmHg; HCO3, 18.1 mmol/L; base excess, −1.2; lactate, 0.33 mmol/L; ionized calcium, 0.88 mmol/L; and potassium, 5.0 mmol/L. Follow-up arterial blood gas analyses were repeated periodically. As we continued to perform CPR, the patient developed severe metabolic acidosis and hypocalcemia characterized by a pH of 6.840, HCO3 of 15.4 mmol/L, and ionized calcium of 0.66 mmol/L. To correct the abnormal findings, we administered appropriate amounts of sodium bicarbonate and calcium chloride. After an hour-long CPR and rewarming, the patient’s esophageal temperature recovered to 33.0°C, which was followed by observing sinus rhythm characterized by an HR of 75 bpm on an ECG monitor and detecting a pulse in the femoral artery with a characteristic ABP waveform that eventually led to the recovery of BP to 130/55 mmHg. Upon confirming recovery of spontaneous circulation (ROSC), the patient was transported to the ICU while we administered dopamine (10 µg/kg/minute), norepinephrine (0.1 µg/kg/minute), and vasopressin (2 U/hour). ABP, HR, and esophageal temperature during anesthesia are shown in Figure 5.
Figure 5. Changes in vital signs (blood pressure (BP), heart rate (HR), and esophageal temperature) over time (5-minute intervals).
bpm, beats per minute.
Cardiac biomarker testing performed after the patient’s transfer to the ICU showed an increased troponin-T level of 1.432 µg/L. Portable Doppler transthoracic echocardiography (TTE) revealed total occlusion of the left main coronary artery, which had undergone PCI previously. Because emergency coronary artery bypass grafting (CABG) was impossible at the time, repeated PCI was considered an alternative treatment option. However, even after his transfer to the ICU, the patient experienced repetitive ventricular fibrillation. Four hours after the patient’s transfer to the ICU, the patient’s guardian requested that CPR should be stopped. Without resuscitation, the patient died.
Discussion
In this case, the patient, who was receiving therapeutic hypothermia, underwent emergency craniectomy owing to worsening brain hemorrhage, and cardiac arrest occurred during general anesthesia. We deduced that the cardiac arrest was owing to the additional decrease in the patient’s core body temperature and myocardial infarction caused by restenosis of the previous PCI-treated lesion.
Therapeutic hypothermia is used to improve neurological outcomes in cardiac arrest, TBI, stroke, and hypoxic–ischemic encephalopathy.1,4 Studies on the efficacy of therapeutic hypothermia for TBI have been less convincing, and they show conflicting results. However, hypothermia is consistently used as an option for TBI because it reduces intracranial pressure and the cerebral metabolic rate, thereby protecting neuronal function in the brain and decreasing the wide range of injuries or reparative processes by attenuating temperature-sensitive mechanisms, such as excitotoxicity, free radical generation, apoptosis, and inflammation.1,2
The cardiovascular effects associated with hypothermia are complex. Hypothermia affects the electrical conduction of myocardial cells, which can induce hypotension or bradycardia and even increase the risk of ventricular fibrillation by prolonging repolarization.4,8 However, mild hypothermia of 32.0°C to 35.0°C demonstrates cardioprotective effects by decreasing myocardial oxygen demand and increasing myocardial perfusion through vasodilation of coronary vessels.9 Therefore, there has been a growing interest in the clinical use of mild therapeutic hypothermia as an adjunctive therapy during ischemic heart disease.10 Our patient received mild therapeutic hypothermia with a target core temperature of 33.0°C, which may have yielded cardioprotective as well as neuroprotective benefits.11
During general anesthesia, hypothermia is induced following an increase in heat loss caused by peripheral vasodilation after the injection of anesthetic drugs, heat redistribution from the core to the periphery, and impaired normal thermoregulatory responses, such as vasoconstriction or shivering.5 In our patient, the cold operating room temperature and irrigating an uncovered surgical area in the brain with cold saline also may have contributed to the additional core body temperature drop in addition to the effects of general anesthesia.12 However, in our case, neurosurgeons requested that anesthesiologists not perform warming during anesthesia. Instead, esophageal temperature and signs that could result from hypothermia, such as ECG changes, were carefully monitored. The esophageal temperature decreased from 33.0°C, which was measured immediately after entering the operation room, to 32.0°C at the time of cardiac arrest. The esophageal temperature further decreased to a minimum of 30.0°C during CPR. With rewarming, the temperature recovered to 33.0°C and finally, ROSC was achieved. Considering that chest compressions, drugs, and defibrillation did not induce ROSC, which was achieved with the recovery of an esophageal temperature of 33.0°C, the additional rapid fall in the patient’s core body temperature during anesthesia may have caused the cardiac arrest.
In this case study, we suspected that restenosis at the site of the previous stent insertion contributed to the cardiac arrest, given the following three findings: First, the patient had a normal preoperative ECG with no signs of ischemia but developed ST-segment depression during anesthesia. Second, repetitive cardiac arrests occurred even after achieving ROSC with rewarming. Third, an increased troponin-T concentration was observed, and suspected re-stenotic occlusion of the left main coronary artery was confirmed by Doppler TTE after the patient’s transfer to the ICU. Accordingly, myocardial infarction was considered the main cause of death.
The mechanisms of myocardial infarction owing to stent restenosis after DES insertion are rapid stent thrombosis or profuse in-stent restenosis (ISR).13 We assume that the lesion characteristics in the left main bifurcation of the coronary artery and DAPT interruption owing to the increasing amount of cerebral hemorrhage contributed to the stent restenosis in this patient. As evidence in support of this assumption, previous studies have shown that stenting at the left main bifurcation and DAPT interruption are strong risk factors for stent thrombosis or ISR.14,15 While controversial, not interrupting aspirin perioperatively could have avoided poor outcomes.16 However, there were no clinical signs related to expected ongoing restenosis of the coronary artery or myocardial ischemia preoperatively in our patient. We suspect that the possible ischemic signs may have been masked by the aforementioned cardioprotective benefits of therapeutic hypothermia. We also believe that the following three conditions contributed to the intraoperative myocardial ischemia and infarction caused by acute restenosis of the coronary stent: First, intraoperative hypotension may have contributed to myocardial injury. The IV anesthetics used during anesthesia, propofol and remifentanil, induced redistribution of blood volume by peripheral vasodilation, resulting in profound hypotension.4 Profound hypotension was strongly associated with myocardial injury and especially harmful for this patient, who had preexisting ischemic heart disease.17 Second, tachycardia, which may have been caused by the inotrope dopamine, used to compensate for the low BP or the stress response to surgical stimulation from shallow anesthetic depth, may have triggered myocardial ischemia by decreasing oxygen delivery and increasing oxygen consumption.17,18 If cardiogenic shock is suspected, as in our case, norepinephrine could be a better option than dopamine.19 Third, the bolus injection of phenylephrine, a selective α1-adrenergic agonist, just before cardiac arrest may have temporarily and abruptly increased cardiac afterload and impaired myocardial perfusion.20
Had we known that coronary artery occlusion would result from hypothermia-induced cardiac arrest and coronary stent restenosis, the treatment of choice would have been CABG surgery while maintaining circulation with emergency cardiopulmonary bypass (CPB) or extracorporeal membrane oxygenation (ECMO) via the femoral artery and vein.11 However, it was difficult to determine the etiology of the cardiac arrest at the time. Even if we had known the cause, ideal treatment would have been impossible, given circumstances beyond our control; all thoracic surgeons were busy operating, and emergency CPB and ECMO were not available. As an alternative, repeat PCI after the patient achieved ROSC could have been performed. However, the patient had a prolonged CPR duration, and he did not maintain the ROSC state for a period sufficient for intervention, which led to a poor outcome.
In this case, it was not possible to accurately determine the reason for the cardiac arrest and for not achieving consistent ROSC. Instead, we conclude that cardiac arrest developed and persisted as a result of the unintentional severe hypothermia that occurred during general anesthesia and/or coronary artery restenosis at the site of the previous PCI. A thorough evaluation of heart-related history is required for TBI patients who have a history of ischemic heart diseases and who undergo therapeutic hypothermia because there may be undiscovered coronary insufficiency owing to the cardioprotective effects of therapeutic hypothermia. In addition, careful monitoring of the patient’s core body temperature, and preventive and countermeasures should be fully prepared and taken against adverse hemodynamic effects, even cardiac arrest, which can occur because of additive hypothermic responses during anesthesia.
Ethics statement: This case report was approved by our Institutional Review Board (EMC 2020-05-005-001), and the case report was performed in compliance with the EQUATOR Network guidelines (the CARE guidelines). All patient details were de-identified, and written consent was obtained from the patient’s guardian for treatment and publication of this report.
Declaration of conflicting interest: The authors declare that there is no conflict of interest.
Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
ORCID iDs: Dong Ho Park https://orcid.org/0000-0002-6587-3756
Tae Woo Kim https://orcid.org/0000-0002-2115-6448
Mo Se Kim https://orcid.org/0000-0001-7362-0030 | 70 kg. | Weight | CC BY-NC | 33499678 | 19,413,557 | 2021-01 |
What was the administration route of drug 'DOPAMINE HYDROCHLORIDE'? | Cardiac arrest caused by accidental severe hypothermia and myocardial infarction during general anesthesia.
Therapeutic hypothermia is often used for traumatic brain injury because of its neuroprotective effect and decreased secondary brain injury. However, this procedure lacks clinical evidence supporting its efficacy, and adverse outcomes have been reported during general anesthesia. A 61-year-old man with a history of percutaneous coronary intervention (PCI) was admitted with traumatic brain injury. Immediately after admission, he underwent mild therapeutic hypothermia with a target temperature of 33.0°C for neuroprotection. During general anesthesia for emergency surgery because he developed a mass effect, hypothermic cardiac arrest occurred following an additional decrease in the core body temperature. Moreover, myocardial infarction caused by restenosis of the previous PCI lesion also contributed to the cardiac arrest. Although the patient recovered spontaneous circulation after an hour-long cardiopulmonary resuscitation with rewarming, he eventually died of subsequent repetitive cardiac arrests. When anesthetizing patients undergoing therapeutic hypothermia, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.
Introduction
Therapeutic hypothermia is an attractive strategy for traumatic brain injury (TBI), which demonstrates neuroprotective effects and decreases secondary brain injury.1 Despite the lack of clinical evidence regarding efficacy, therapeutic hypothermia after TBI has been used for at least half a century because of its potential benefits, which are supported by numerous preclinical studies.2,3 However, there have been no reports on the danger of unintentional severe hypothermia during general anesthesia caused by the additional drop in temperature of patients undergoing therapeutic hypothermia. When anesthetizing patients with comorbidities, such as ischemic heart disease, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.4,5
Case report
A 61-year-old man (height, 168 cm; weight, 70 kg) presented to our emergency room (ER) with head trauma after falling on a cement floor from a 5-meter-tall construction site. Upon arrival at the ER, the patient had a Glasgow coma scale score of 11 and the following vital signs: blood pressure (BP), 147/83 mmHg; heart rate (HR), 86 beats per minute (bpm); respiratory rate, 24 breaths per minute; body temperature from the auditory canal, 36.8°C; and percutaneous oxygen saturation (SpO2), 97%. Upon discovering signs of acute subdural hemorrhage, subarachnoid hemorrhage, and intraventricular hemorrhage from a brain computed tomography (CT) scan performed in the ER, we admitted the patient to the intensive care unit (ICU) for close monitoring. After approximately 6 hours, we performed follow-up brain CT because the patient’s mental status had deteriorated from drowsy to stupor. Upon observing a mass effect characterized by an increase in the hemorrhage volume and a left midline shift on the CT images, the patient was scheduled for immediate intubation and emergency decompressive craniectomy (Figure 1).
Figure 1. Brain computed tomography (CT) images taken 6 hours after intensive care unit admission. (a) Axial view showing acute subarachnoid hemorrhage in the peri-mesencephalic cistern (arrow A) and subdural hemorrhage in the right inferior temporal (arrow B) and right parietal (arrow C) regions. (b) Axial view showing acute subdural hemorrhage in the right temporal region (black arrow) and left midline shift.
Although the patient had a history of ischemic heart disease with percutaneous coronary intervention (PCI), it was difficult to obtain accurate information because there was no relevant data in the patient’s medical records from the authors’ hospital, and his drowsy state interfered with detailed interviews. According to the chest CT scan performed immediately after admission, coronary stents had been inserted at the bifurcation site of the left main coronary artery (Figure 2). It also appeared that the inserted stents were drug-eluting stents (DES) considering that he had received dual antiplatelet therapy (DAPT) with aspirin and clopidogrel (Plavix®; Sanofi, Bridgewater, NJ, USA) a switching platelet P2Y12 inhibitor, for approximately 4 months. However, DAPT was discontinued owing to the worsening cerebral hemorrhage after hospitalization; bridging antiplatelet therapy was also not performed. Before the surgery, the patient showed a normal electrocardiogram (ECG), and cardiac biomarker testing revealed a troponin-T concentration of 0.006 µg/L, which was within normal limits (Figure 3). The patient demonstrated satisfactory functional capacity to be able to work at a construction site just before the trauma. Therefore, an additional preoperative cardiac evaluation was not considered necessary.
Figure 2. Chest computed tomography (CT) images showing coronary stents in the main bifurcation of the left coronary arteries (white arrow in (a) and (b)).
Figure 3. Preoperative electrocardiogram of the patient showing normal sinus rhythm with a heart rate of 95 bpm.
bpm, beats per minute.
Immediately after admission to the ICU, the patient’s bladder temperature was 38.0°C. Using a hydrogel-coated water-circulating pad (Arctic Sun®; Medivance Inc., Louisville, CO, USA), which is a surface cooling device, the patient underwent mild therapeutic hypothermia with a target temperature of 33.0°C. During therapeutic hypothermia, sedation and shivering control were performed using continuous rate infusions of midazolam at 5 mg/hour and remifentanil at 0.2 µg/kg/minute. The patient’s bladder temperature was 33.4°C just before he was transferred to the operating room. Preoperative laboratory tests showed no unusual findings except for mildly elevated aspartate aminotransferase (73 U/L) and alanine aminotransferase (51 U/L) concentrations.
The patient’s vital signs upon arrival at the operating room were a BP of 130/70 mmHg and SpO2 of 100%. Lead II of the ECG was monitored, and there were no unusual findings except sinus tachycardia with an HR of 100 bpm on the initial ECG. Arterial BP (ABP), central venous pressure (CVP), and esophageal temperature were monitored from the right radial artery, right subclavian vein, and lower third of the esophagus using a temperature sensor attached to the esophageal stethoscope, respectively. The initial ABP, CVP, and esophageal temperature were 140/85 mmHg, 6 mmHg, and 33.0°C, respectively. Arterial blood gas analysis was performed in the arterial line, with values calculated as for a body temperature of 37°C, which revealed the following: pH, 7.400; partial pressure of carbon dioxide (PCO2), 40 mmHg; partial pressure of oxygen (PO2), 174 mmHg; bicarbonate (HCO3), 24.8 mmol/L; base excess, 0.0; lactate, 0.17 mmol/L; ionized calcium, 1.04 mmol/L; and potassium, 3.3 mmol/L. Total intravenous (IV) anesthesia was performed using a target-controlled infusion of propofol and remifentanil in accordance with the Schnider6 and Minto7 models, respectively. Propofol and remifentanil were infused continuously with effect-site concentrations (Ce) of 1.5 µg/mL and 1.5 ng/mL, respectively. Concurrently, 50 mg of rocuronium was administered by IV bolus injection for neuromuscular blockade. Mechanical ventilation was adjusted to maintain the end-tidal carbon dioxide value at 30 to 33 mmHg.
As ABP decreased to 78/42 mmHg and CVP decreased to 4 mmHg within 10 minutes after anesthesia, the Ce of propofol and remifentanil were lowered to 0.8 µg/mL and 1.0 ng/mL, respectively, and approximately 1 L of balanced crystalloid fluid loading was performed. However, the patient’s BP did not recover, and a dopamine continuous infusion (5 µg/kg/minute) was initiated. When a surgical incision was made 10 minutes after beginning the dopamine infusion, the patient’s ABP and HR increased to 115/70 mmHg and 120 bpm, respectively, and marked ST-segment depression (−2.4 mm) was found, which was not observed on the preoperative or initial ECGs (Figure 4). Subsequently, the depth of anesthesia was increased by increasing the Ce of propofol to 1.0 to 1.2 µg/mL, followed by a continuous infusion of isosorbide dinitrate (1 µg/kg/minute). While HR was maintained at 110 to 120 bpm, ABP gradually decreased to 70/40 mmHg after approximately 5 minutes. To treat the hypotension, isosorbide dinitrate was discontinued, and a bolus injection of 100 µg phenylephrine was administered, after which, ABP increased temporarily to 95/60 mmHg. Subsequently, HR decreased suddenly and rapidly, and ventricular fibrillation occurred. The anesthesiologists immediately discontinued the surgery and performed cardiopulmonary resuscitation (CPR). The patient’s esophageal temperature at the time was 32.0°C.
Figure 4. Intraoperative electrocardiogram of the patient showing sinus tachycardia with significant ST depression. The ST depression on the monitor represented approximately −2.4 at the time.
Chest compressions were performed with periodic defibrillation with 200 J, and epinephrine injections were administered every 3 minutes; these measures elicited no response. During CPR, ECG monitoring showed temporary severe bradycardia (HR, 10–30 bpm). However, this was considered pulseless electrical activity because no pulse was detected in the patient’s femoral artery. Soon afterward, there was a rapid transition to ventricular fibrillation, and CPR was repeated. At the time, the patient’s esophageal temperature dropped to 30.0°C. According to the refractory state after the defibrillation and drug injections, we considered that the cardiac arrest was caused by hypothermia. To reverse the cause, a forced-air warming device and a warming blanket (Bair Hugger®; Arizant Healthcare Inc., Eden Prairie, MN, USA) were used for surface warming, the IV fluids were warmed, and surgical wound closure and dressing were performed rapidly to prevent surface heat loss. The first arterial blood gas analysis after ventricular fibrillation, calculated as for a body temperature of 37°C, showed the following: pH, 7.690; PCO2, 15 mmHg; PO2, 260 mmHg; HCO3, 18.1 mmol/L; base excess, −1.2; lactate, 0.33 mmol/L; ionized calcium, 0.88 mmol/L; and potassium, 5.0 mmol/L. Follow-up arterial blood gas analyses were repeated periodically. As we continued to perform CPR, the patient developed severe metabolic acidosis and hypocalcemia characterized by a pH of 6.840, HCO3 of 15.4 mmol/L, and ionized calcium of 0.66 mmol/L. To correct the abnormal findings, we administered appropriate amounts of sodium bicarbonate and calcium chloride. After an hour-long CPR and rewarming, the patient’s esophageal temperature recovered to 33.0°C, which was followed by observing sinus rhythm characterized by an HR of 75 bpm on an ECG monitor and detecting a pulse in the femoral artery with a characteristic ABP waveform that eventually led to the recovery of BP to 130/55 mmHg. Upon confirming recovery of spontaneous circulation (ROSC), the patient was transported to the ICU while we administered dopamine (10 µg/kg/minute), norepinephrine (0.1 µg/kg/minute), and vasopressin (2 U/hour). ABP, HR, and esophageal temperature during anesthesia are shown in Figure 5.
Figure 5. Changes in vital signs (blood pressure (BP), heart rate (HR), and esophageal temperature) over time (5-minute intervals).
bpm, beats per minute.
Cardiac biomarker testing performed after the patient’s transfer to the ICU showed an increased troponin-T level of 1.432 µg/L. Portable Doppler transthoracic echocardiography (TTE) revealed total occlusion of the left main coronary artery, which had undergone PCI previously. Because emergency coronary artery bypass grafting (CABG) was impossible at the time, repeated PCI was considered an alternative treatment option. However, even after his transfer to the ICU, the patient experienced repetitive ventricular fibrillation. Four hours after the patient’s transfer to the ICU, the patient’s guardian requested that CPR should be stopped. Without resuscitation, the patient died.
Discussion
In this case, the patient, who was receiving therapeutic hypothermia, underwent emergency craniectomy owing to worsening brain hemorrhage, and cardiac arrest occurred during general anesthesia. We deduced that the cardiac arrest was owing to the additional decrease in the patient’s core body temperature and myocardial infarction caused by restenosis of the previous PCI-treated lesion.
Therapeutic hypothermia is used to improve neurological outcomes in cardiac arrest, TBI, stroke, and hypoxic–ischemic encephalopathy.1,4 Studies on the efficacy of therapeutic hypothermia for TBI have been less convincing, and they show conflicting results. However, hypothermia is consistently used as an option for TBI because it reduces intracranial pressure and the cerebral metabolic rate, thereby protecting neuronal function in the brain and decreasing the wide range of injuries or reparative processes by attenuating temperature-sensitive mechanisms, such as excitotoxicity, free radical generation, apoptosis, and inflammation.1,2
The cardiovascular effects associated with hypothermia are complex. Hypothermia affects the electrical conduction of myocardial cells, which can induce hypotension or bradycardia and even increase the risk of ventricular fibrillation by prolonging repolarization.4,8 However, mild hypothermia of 32.0°C to 35.0°C demonstrates cardioprotective effects by decreasing myocardial oxygen demand and increasing myocardial perfusion through vasodilation of coronary vessels.9 Therefore, there has been a growing interest in the clinical use of mild therapeutic hypothermia as an adjunctive therapy during ischemic heart disease.10 Our patient received mild therapeutic hypothermia with a target core temperature of 33.0°C, which may have yielded cardioprotective as well as neuroprotective benefits.11
During general anesthesia, hypothermia is induced following an increase in heat loss caused by peripheral vasodilation after the injection of anesthetic drugs, heat redistribution from the core to the periphery, and impaired normal thermoregulatory responses, such as vasoconstriction or shivering.5 In our patient, the cold operating room temperature and irrigating an uncovered surgical area in the brain with cold saline also may have contributed to the additional core body temperature drop in addition to the effects of general anesthesia.12 However, in our case, neurosurgeons requested that anesthesiologists not perform warming during anesthesia. Instead, esophageal temperature and signs that could result from hypothermia, such as ECG changes, were carefully monitored. The esophageal temperature decreased from 33.0°C, which was measured immediately after entering the operation room, to 32.0°C at the time of cardiac arrest. The esophageal temperature further decreased to a minimum of 30.0°C during CPR. With rewarming, the temperature recovered to 33.0°C and finally, ROSC was achieved. Considering that chest compressions, drugs, and defibrillation did not induce ROSC, which was achieved with the recovery of an esophageal temperature of 33.0°C, the additional rapid fall in the patient’s core body temperature during anesthesia may have caused the cardiac arrest.
In this case study, we suspected that restenosis at the site of the previous stent insertion contributed to the cardiac arrest, given the following three findings: First, the patient had a normal preoperative ECG with no signs of ischemia but developed ST-segment depression during anesthesia. Second, repetitive cardiac arrests occurred even after achieving ROSC with rewarming. Third, an increased troponin-T concentration was observed, and suspected re-stenotic occlusion of the left main coronary artery was confirmed by Doppler TTE after the patient’s transfer to the ICU. Accordingly, myocardial infarction was considered the main cause of death.
The mechanisms of myocardial infarction owing to stent restenosis after DES insertion are rapid stent thrombosis or profuse in-stent restenosis (ISR).13 We assume that the lesion characteristics in the left main bifurcation of the coronary artery and DAPT interruption owing to the increasing amount of cerebral hemorrhage contributed to the stent restenosis in this patient. As evidence in support of this assumption, previous studies have shown that stenting at the left main bifurcation and DAPT interruption are strong risk factors for stent thrombosis or ISR.14,15 While controversial, not interrupting aspirin perioperatively could have avoided poor outcomes.16 However, there were no clinical signs related to expected ongoing restenosis of the coronary artery or myocardial ischemia preoperatively in our patient. We suspect that the possible ischemic signs may have been masked by the aforementioned cardioprotective benefits of therapeutic hypothermia. We also believe that the following three conditions contributed to the intraoperative myocardial ischemia and infarction caused by acute restenosis of the coronary stent: First, intraoperative hypotension may have contributed to myocardial injury. The IV anesthetics used during anesthesia, propofol and remifentanil, induced redistribution of blood volume by peripheral vasodilation, resulting in profound hypotension.4 Profound hypotension was strongly associated with myocardial injury and especially harmful for this patient, who had preexisting ischemic heart disease.17 Second, tachycardia, which may have been caused by the inotrope dopamine, used to compensate for the low BP or the stress response to surgical stimulation from shallow anesthetic depth, may have triggered myocardial ischemia by decreasing oxygen delivery and increasing oxygen consumption.17,18 If cardiogenic shock is suspected, as in our case, norepinephrine could be a better option than dopamine.19 Third, the bolus injection of phenylephrine, a selective α1-adrenergic agonist, just before cardiac arrest may have temporarily and abruptly increased cardiac afterload and impaired myocardial perfusion.20
Had we known that coronary artery occlusion would result from hypothermia-induced cardiac arrest and coronary stent restenosis, the treatment of choice would have been CABG surgery while maintaining circulation with emergency cardiopulmonary bypass (CPB) or extracorporeal membrane oxygenation (ECMO) via the femoral artery and vein.11 However, it was difficult to determine the etiology of the cardiac arrest at the time. Even if we had known the cause, ideal treatment would have been impossible, given circumstances beyond our control; all thoracic surgeons were busy operating, and emergency CPB and ECMO were not available. As an alternative, repeat PCI after the patient achieved ROSC could have been performed. However, the patient had a prolonged CPR duration, and he did not maintain the ROSC state for a period sufficient for intervention, which led to a poor outcome.
In this case, it was not possible to accurately determine the reason for the cardiac arrest and for not achieving consistent ROSC. Instead, we conclude that cardiac arrest developed and persisted as a result of the unintentional severe hypothermia that occurred during general anesthesia and/or coronary artery restenosis at the site of the previous PCI. A thorough evaluation of heart-related history is required for TBI patients who have a history of ischemic heart diseases and who undergo therapeutic hypothermia because there may be undiscovered coronary insufficiency owing to the cardioprotective effects of therapeutic hypothermia. In addition, careful monitoring of the patient’s core body temperature, and preventive and countermeasures should be fully prepared and taken against adverse hemodynamic effects, even cardiac arrest, which can occur because of additive hypothermic responses during anesthesia.
Ethics statement: This case report was approved by our Institutional Review Board (EMC 2020-05-005-001), and the case report was performed in compliance with the EQUATOR Network guidelines (the CARE guidelines). All patient details were de-identified, and written consent was obtained from the patient’s guardian for treatment and publication of this report.
Declaration of conflicting interest: The authors declare that there is no conflict of interest.
Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
ORCID iDs: Dong Ho Park https://orcid.org/0000-0002-6587-3756
Tae Woo Kim https://orcid.org/0000-0002-2115-6448
Mo Se Kim https://orcid.org/0000-0001-7362-0030 | Intravenous drip | DrugAdministrationRoute | CC BY-NC | 33499678 | 19,429,456 | 2021-01 |
What was the administration route of drug 'MIDAZOLAM'? | Cardiac arrest caused by accidental severe hypothermia and myocardial infarction during general anesthesia.
Therapeutic hypothermia is often used for traumatic brain injury because of its neuroprotective effect and decreased secondary brain injury. However, this procedure lacks clinical evidence supporting its efficacy, and adverse outcomes have been reported during general anesthesia. A 61-year-old man with a history of percutaneous coronary intervention (PCI) was admitted with traumatic brain injury. Immediately after admission, he underwent mild therapeutic hypothermia with a target temperature of 33.0°C for neuroprotection. During general anesthesia for emergency surgery because he developed a mass effect, hypothermic cardiac arrest occurred following an additional decrease in the core body temperature. Moreover, myocardial infarction caused by restenosis of the previous PCI lesion also contributed to the cardiac arrest. Although the patient recovered spontaneous circulation after an hour-long cardiopulmonary resuscitation with rewarming, he eventually died of subsequent repetitive cardiac arrests. When anesthetizing patients undergoing therapeutic hypothermia, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.
Introduction
Therapeutic hypothermia is an attractive strategy for traumatic brain injury (TBI), which demonstrates neuroprotective effects and decreases secondary brain injury.1 Despite the lack of clinical evidence regarding efficacy, therapeutic hypothermia after TBI has been used for at least half a century because of its potential benefits, which are supported by numerous preclinical studies.2,3 However, there have been no reports on the danger of unintentional severe hypothermia during general anesthesia caused by the additional drop in temperature of patients undergoing therapeutic hypothermia. When anesthetizing patients with comorbidities, such as ischemic heart disease, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.4,5
Case report
A 61-year-old man (height, 168 cm; weight, 70 kg) presented to our emergency room (ER) with head trauma after falling on a cement floor from a 5-meter-tall construction site. Upon arrival at the ER, the patient had a Glasgow coma scale score of 11 and the following vital signs: blood pressure (BP), 147/83 mmHg; heart rate (HR), 86 beats per minute (bpm); respiratory rate, 24 breaths per minute; body temperature from the auditory canal, 36.8°C; and percutaneous oxygen saturation (SpO2), 97%. Upon discovering signs of acute subdural hemorrhage, subarachnoid hemorrhage, and intraventricular hemorrhage from a brain computed tomography (CT) scan performed in the ER, we admitted the patient to the intensive care unit (ICU) for close monitoring. After approximately 6 hours, we performed follow-up brain CT because the patient’s mental status had deteriorated from drowsy to stupor. Upon observing a mass effect characterized by an increase in the hemorrhage volume and a left midline shift on the CT images, the patient was scheduled for immediate intubation and emergency decompressive craniectomy (Figure 1).
Figure 1. Brain computed tomography (CT) images taken 6 hours after intensive care unit admission. (a) Axial view showing acute subarachnoid hemorrhage in the peri-mesencephalic cistern (arrow A) and subdural hemorrhage in the right inferior temporal (arrow B) and right parietal (arrow C) regions. (b) Axial view showing acute subdural hemorrhage in the right temporal region (black arrow) and left midline shift.
Although the patient had a history of ischemic heart disease with percutaneous coronary intervention (PCI), it was difficult to obtain accurate information because there was no relevant data in the patient’s medical records from the authors’ hospital, and his drowsy state interfered with detailed interviews. According to the chest CT scan performed immediately after admission, coronary stents had been inserted at the bifurcation site of the left main coronary artery (Figure 2). It also appeared that the inserted stents were drug-eluting stents (DES) considering that he had received dual antiplatelet therapy (DAPT) with aspirin and clopidogrel (Plavix®; Sanofi, Bridgewater, NJ, USA) a switching platelet P2Y12 inhibitor, for approximately 4 months. However, DAPT was discontinued owing to the worsening cerebral hemorrhage after hospitalization; bridging antiplatelet therapy was also not performed. Before the surgery, the patient showed a normal electrocardiogram (ECG), and cardiac biomarker testing revealed a troponin-T concentration of 0.006 µg/L, which was within normal limits (Figure 3). The patient demonstrated satisfactory functional capacity to be able to work at a construction site just before the trauma. Therefore, an additional preoperative cardiac evaluation was not considered necessary.
Figure 2. Chest computed tomography (CT) images showing coronary stents in the main bifurcation of the left coronary arteries (white arrow in (a) and (b)).
Figure 3. Preoperative electrocardiogram of the patient showing normal sinus rhythm with a heart rate of 95 bpm.
bpm, beats per minute.
Immediately after admission to the ICU, the patient’s bladder temperature was 38.0°C. Using a hydrogel-coated water-circulating pad (Arctic Sun®; Medivance Inc., Louisville, CO, USA), which is a surface cooling device, the patient underwent mild therapeutic hypothermia with a target temperature of 33.0°C. During therapeutic hypothermia, sedation and shivering control were performed using continuous rate infusions of midazolam at 5 mg/hour and remifentanil at 0.2 µg/kg/minute. The patient’s bladder temperature was 33.4°C just before he was transferred to the operating room. Preoperative laboratory tests showed no unusual findings except for mildly elevated aspartate aminotransferase (73 U/L) and alanine aminotransferase (51 U/L) concentrations.
The patient’s vital signs upon arrival at the operating room were a BP of 130/70 mmHg and SpO2 of 100%. Lead II of the ECG was monitored, and there were no unusual findings except sinus tachycardia with an HR of 100 bpm on the initial ECG. Arterial BP (ABP), central venous pressure (CVP), and esophageal temperature were monitored from the right radial artery, right subclavian vein, and lower third of the esophagus using a temperature sensor attached to the esophageal stethoscope, respectively. The initial ABP, CVP, and esophageal temperature were 140/85 mmHg, 6 mmHg, and 33.0°C, respectively. Arterial blood gas analysis was performed in the arterial line, with values calculated as for a body temperature of 37°C, which revealed the following: pH, 7.400; partial pressure of carbon dioxide (PCO2), 40 mmHg; partial pressure of oxygen (PO2), 174 mmHg; bicarbonate (HCO3), 24.8 mmol/L; base excess, 0.0; lactate, 0.17 mmol/L; ionized calcium, 1.04 mmol/L; and potassium, 3.3 mmol/L. Total intravenous (IV) anesthesia was performed using a target-controlled infusion of propofol and remifentanil in accordance with the Schnider6 and Minto7 models, respectively. Propofol and remifentanil were infused continuously with effect-site concentrations (Ce) of 1.5 µg/mL and 1.5 ng/mL, respectively. Concurrently, 50 mg of rocuronium was administered by IV bolus injection for neuromuscular blockade. Mechanical ventilation was adjusted to maintain the end-tidal carbon dioxide value at 30 to 33 mmHg.
As ABP decreased to 78/42 mmHg and CVP decreased to 4 mmHg within 10 minutes after anesthesia, the Ce of propofol and remifentanil were lowered to 0.8 µg/mL and 1.0 ng/mL, respectively, and approximately 1 L of balanced crystalloid fluid loading was performed. However, the patient’s BP did not recover, and a dopamine continuous infusion (5 µg/kg/minute) was initiated. When a surgical incision was made 10 minutes after beginning the dopamine infusion, the patient’s ABP and HR increased to 115/70 mmHg and 120 bpm, respectively, and marked ST-segment depression (−2.4 mm) was found, which was not observed on the preoperative or initial ECGs (Figure 4). Subsequently, the depth of anesthesia was increased by increasing the Ce of propofol to 1.0 to 1.2 µg/mL, followed by a continuous infusion of isosorbide dinitrate (1 µg/kg/minute). While HR was maintained at 110 to 120 bpm, ABP gradually decreased to 70/40 mmHg after approximately 5 minutes. To treat the hypotension, isosorbide dinitrate was discontinued, and a bolus injection of 100 µg phenylephrine was administered, after which, ABP increased temporarily to 95/60 mmHg. Subsequently, HR decreased suddenly and rapidly, and ventricular fibrillation occurred. The anesthesiologists immediately discontinued the surgery and performed cardiopulmonary resuscitation (CPR). The patient’s esophageal temperature at the time was 32.0°C.
Figure 4. Intraoperative electrocardiogram of the patient showing sinus tachycardia with significant ST depression. The ST depression on the monitor represented approximately −2.4 at the time.
Chest compressions were performed with periodic defibrillation with 200 J, and epinephrine injections were administered every 3 minutes; these measures elicited no response. During CPR, ECG monitoring showed temporary severe bradycardia (HR, 10–30 bpm). However, this was considered pulseless electrical activity because no pulse was detected in the patient’s femoral artery. Soon afterward, there was a rapid transition to ventricular fibrillation, and CPR was repeated. At the time, the patient’s esophageal temperature dropped to 30.0°C. According to the refractory state after the defibrillation and drug injections, we considered that the cardiac arrest was caused by hypothermia. To reverse the cause, a forced-air warming device and a warming blanket (Bair Hugger®; Arizant Healthcare Inc., Eden Prairie, MN, USA) were used for surface warming, the IV fluids were warmed, and surgical wound closure and dressing were performed rapidly to prevent surface heat loss. The first arterial blood gas analysis after ventricular fibrillation, calculated as for a body temperature of 37°C, showed the following: pH, 7.690; PCO2, 15 mmHg; PO2, 260 mmHg; HCO3, 18.1 mmol/L; base excess, −1.2; lactate, 0.33 mmol/L; ionized calcium, 0.88 mmol/L; and potassium, 5.0 mmol/L. Follow-up arterial blood gas analyses were repeated periodically. As we continued to perform CPR, the patient developed severe metabolic acidosis and hypocalcemia characterized by a pH of 6.840, HCO3 of 15.4 mmol/L, and ionized calcium of 0.66 mmol/L. To correct the abnormal findings, we administered appropriate amounts of sodium bicarbonate and calcium chloride. After an hour-long CPR and rewarming, the patient’s esophageal temperature recovered to 33.0°C, which was followed by observing sinus rhythm characterized by an HR of 75 bpm on an ECG monitor and detecting a pulse in the femoral artery with a characteristic ABP waveform that eventually led to the recovery of BP to 130/55 mmHg. Upon confirming recovery of spontaneous circulation (ROSC), the patient was transported to the ICU while we administered dopamine (10 µg/kg/minute), norepinephrine (0.1 µg/kg/minute), and vasopressin (2 U/hour). ABP, HR, and esophageal temperature during anesthesia are shown in Figure 5.
Figure 5. Changes in vital signs (blood pressure (BP), heart rate (HR), and esophageal temperature) over time (5-minute intervals).
bpm, beats per minute.
Cardiac biomarker testing performed after the patient’s transfer to the ICU showed an increased troponin-T level of 1.432 µg/L. Portable Doppler transthoracic echocardiography (TTE) revealed total occlusion of the left main coronary artery, which had undergone PCI previously. Because emergency coronary artery bypass grafting (CABG) was impossible at the time, repeated PCI was considered an alternative treatment option. However, even after his transfer to the ICU, the patient experienced repetitive ventricular fibrillation. Four hours after the patient’s transfer to the ICU, the patient’s guardian requested that CPR should be stopped. Without resuscitation, the patient died.
Discussion
In this case, the patient, who was receiving therapeutic hypothermia, underwent emergency craniectomy owing to worsening brain hemorrhage, and cardiac arrest occurred during general anesthesia. We deduced that the cardiac arrest was owing to the additional decrease in the patient’s core body temperature and myocardial infarction caused by restenosis of the previous PCI-treated lesion.
Therapeutic hypothermia is used to improve neurological outcomes in cardiac arrest, TBI, stroke, and hypoxic–ischemic encephalopathy.1,4 Studies on the efficacy of therapeutic hypothermia for TBI have been less convincing, and they show conflicting results. However, hypothermia is consistently used as an option for TBI because it reduces intracranial pressure and the cerebral metabolic rate, thereby protecting neuronal function in the brain and decreasing the wide range of injuries or reparative processes by attenuating temperature-sensitive mechanisms, such as excitotoxicity, free radical generation, apoptosis, and inflammation.1,2
The cardiovascular effects associated with hypothermia are complex. Hypothermia affects the electrical conduction of myocardial cells, which can induce hypotension or bradycardia and even increase the risk of ventricular fibrillation by prolonging repolarization.4,8 However, mild hypothermia of 32.0°C to 35.0°C demonstrates cardioprotective effects by decreasing myocardial oxygen demand and increasing myocardial perfusion through vasodilation of coronary vessels.9 Therefore, there has been a growing interest in the clinical use of mild therapeutic hypothermia as an adjunctive therapy during ischemic heart disease.10 Our patient received mild therapeutic hypothermia with a target core temperature of 33.0°C, which may have yielded cardioprotective as well as neuroprotective benefits.11
During general anesthesia, hypothermia is induced following an increase in heat loss caused by peripheral vasodilation after the injection of anesthetic drugs, heat redistribution from the core to the periphery, and impaired normal thermoregulatory responses, such as vasoconstriction or shivering.5 In our patient, the cold operating room temperature and irrigating an uncovered surgical area in the brain with cold saline also may have contributed to the additional core body temperature drop in addition to the effects of general anesthesia.12 However, in our case, neurosurgeons requested that anesthesiologists not perform warming during anesthesia. Instead, esophageal temperature and signs that could result from hypothermia, such as ECG changes, were carefully monitored. The esophageal temperature decreased from 33.0°C, which was measured immediately after entering the operation room, to 32.0°C at the time of cardiac arrest. The esophageal temperature further decreased to a minimum of 30.0°C during CPR. With rewarming, the temperature recovered to 33.0°C and finally, ROSC was achieved. Considering that chest compressions, drugs, and defibrillation did not induce ROSC, which was achieved with the recovery of an esophageal temperature of 33.0°C, the additional rapid fall in the patient’s core body temperature during anesthesia may have caused the cardiac arrest.
In this case study, we suspected that restenosis at the site of the previous stent insertion contributed to the cardiac arrest, given the following three findings: First, the patient had a normal preoperative ECG with no signs of ischemia but developed ST-segment depression during anesthesia. Second, repetitive cardiac arrests occurred even after achieving ROSC with rewarming. Third, an increased troponin-T concentration was observed, and suspected re-stenotic occlusion of the left main coronary artery was confirmed by Doppler TTE after the patient’s transfer to the ICU. Accordingly, myocardial infarction was considered the main cause of death.
The mechanisms of myocardial infarction owing to stent restenosis after DES insertion are rapid stent thrombosis or profuse in-stent restenosis (ISR).13 We assume that the lesion characteristics in the left main bifurcation of the coronary artery and DAPT interruption owing to the increasing amount of cerebral hemorrhage contributed to the stent restenosis in this patient. As evidence in support of this assumption, previous studies have shown that stenting at the left main bifurcation and DAPT interruption are strong risk factors for stent thrombosis or ISR.14,15 While controversial, not interrupting aspirin perioperatively could have avoided poor outcomes.16 However, there were no clinical signs related to expected ongoing restenosis of the coronary artery or myocardial ischemia preoperatively in our patient. We suspect that the possible ischemic signs may have been masked by the aforementioned cardioprotective benefits of therapeutic hypothermia. We also believe that the following three conditions contributed to the intraoperative myocardial ischemia and infarction caused by acute restenosis of the coronary stent: First, intraoperative hypotension may have contributed to myocardial injury. The IV anesthetics used during anesthesia, propofol and remifentanil, induced redistribution of blood volume by peripheral vasodilation, resulting in profound hypotension.4 Profound hypotension was strongly associated with myocardial injury and especially harmful for this patient, who had preexisting ischemic heart disease.17 Second, tachycardia, which may have been caused by the inotrope dopamine, used to compensate for the low BP or the stress response to surgical stimulation from shallow anesthetic depth, may have triggered myocardial ischemia by decreasing oxygen delivery and increasing oxygen consumption.17,18 If cardiogenic shock is suspected, as in our case, norepinephrine could be a better option than dopamine.19 Third, the bolus injection of phenylephrine, a selective α1-adrenergic agonist, just before cardiac arrest may have temporarily and abruptly increased cardiac afterload and impaired myocardial perfusion.20
Had we known that coronary artery occlusion would result from hypothermia-induced cardiac arrest and coronary stent restenosis, the treatment of choice would have been CABG surgery while maintaining circulation with emergency cardiopulmonary bypass (CPB) or extracorporeal membrane oxygenation (ECMO) via the femoral artery and vein.11 However, it was difficult to determine the etiology of the cardiac arrest at the time. Even if we had known the cause, ideal treatment would have been impossible, given circumstances beyond our control; all thoracic surgeons were busy operating, and emergency CPB and ECMO were not available. As an alternative, repeat PCI after the patient achieved ROSC could have been performed. However, the patient had a prolonged CPR duration, and he did not maintain the ROSC state for a period sufficient for intervention, which led to a poor outcome.
In this case, it was not possible to accurately determine the reason for the cardiac arrest and for not achieving consistent ROSC. Instead, we conclude that cardiac arrest developed and persisted as a result of the unintentional severe hypothermia that occurred during general anesthesia and/or coronary artery restenosis at the site of the previous PCI. A thorough evaluation of heart-related history is required for TBI patients who have a history of ischemic heart diseases and who undergo therapeutic hypothermia because there may be undiscovered coronary insufficiency owing to the cardioprotective effects of therapeutic hypothermia. In addition, careful monitoring of the patient’s core body temperature, and preventive and countermeasures should be fully prepared and taken against adverse hemodynamic effects, even cardiac arrest, which can occur because of additive hypothermic responses during anesthesia.
Ethics statement: This case report was approved by our Institutional Review Board (EMC 2020-05-005-001), and the case report was performed in compliance with the EQUATOR Network guidelines (the CARE guidelines). All patient details were de-identified, and written consent was obtained from the patient’s guardian for treatment and publication of this report.
Declaration of conflicting interest: The authors declare that there is no conflict of interest.
Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
ORCID iDs: Dong Ho Park https://orcid.org/0000-0002-6587-3756
Tae Woo Kim https://orcid.org/0000-0002-2115-6448
Mo Se Kim https://orcid.org/0000-0001-7362-0030 | Intravenous drip | DrugAdministrationRoute | CC BY-NC | 33499678 | 19,429,456 | 2021-01 |
What was the administration route of drug 'PHENYLEPHRINE HYDROCHLORIDE'? | Cardiac arrest caused by accidental severe hypothermia and myocardial infarction during general anesthesia.
Therapeutic hypothermia is often used for traumatic brain injury because of its neuroprotective effect and decreased secondary brain injury. However, this procedure lacks clinical evidence supporting its efficacy, and adverse outcomes have been reported during general anesthesia. A 61-year-old man with a history of percutaneous coronary intervention (PCI) was admitted with traumatic brain injury. Immediately after admission, he underwent mild therapeutic hypothermia with a target temperature of 33.0°C for neuroprotection. During general anesthesia for emergency surgery because he developed a mass effect, hypothermic cardiac arrest occurred following an additional decrease in the core body temperature. Moreover, myocardial infarction caused by restenosis of the previous PCI lesion also contributed to the cardiac arrest. Although the patient recovered spontaneous circulation after an hour-long cardiopulmonary resuscitation with rewarming, he eventually died of subsequent repetitive cardiac arrests. When anesthetizing patients undergoing therapeutic hypothermia, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.
Introduction
Therapeutic hypothermia is an attractive strategy for traumatic brain injury (TBI), which demonstrates neuroprotective effects and decreases secondary brain injury.1 Despite the lack of clinical evidence regarding efficacy, therapeutic hypothermia after TBI has been used for at least half a century because of its potential benefits, which are supported by numerous preclinical studies.2,3 However, there have been no reports on the danger of unintentional severe hypothermia during general anesthesia caused by the additional drop in temperature of patients undergoing therapeutic hypothermia. When anesthetizing patients with comorbidities, such as ischemic heart disease, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.4,5
Case report
A 61-year-old man (height, 168 cm; weight, 70 kg) presented to our emergency room (ER) with head trauma after falling on a cement floor from a 5-meter-tall construction site. Upon arrival at the ER, the patient had a Glasgow coma scale score of 11 and the following vital signs: blood pressure (BP), 147/83 mmHg; heart rate (HR), 86 beats per minute (bpm); respiratory rate, 24 breaths per minute; body temperature from the auditory canal, 36.8°C; and percutaneous oxygen saturation (SpO2), 97%. Upon discovering signs of acute subdural hemorrhage, subarachnoid hemorrhage, and intraventricular hemorrhage from a brain computed tomography (CT) scan performed in the ER, we admitted the patient to the intensive care unit (ICU) for close monitoring. After approximately 6 hours, we performed follow-up brain CT because the patient’s mental status had deteriorated from drowsy to stupor. Upon observing a mass effect characterized by an increase in the hemorrhage volume and a left midline shift on the CT images, the patient was scheduled for immediate intubation and emergency decompressive craniectomy (Figure 1).
Figure 1. Brain computed tomography (CT) images taken 6 hours after intensive care unit admission. (a) Axial view showing acute subarachnoid hemorrhage in the peri-mesencephalic cistern (arrow A) and subdural hemorrhage in the right inferior temporal (arrow B) and right parietal (arrow C) regions. (b) Axial view showing acute subdural hemorrhage in the right temporal region (black arrow) and left midline shift.
Although the patient had a history of ischemic heart disease with percutaneous coronary intervention (PCI), it was difficult to obtain accurate information because there was no relevant data in the patient’s medical records from the authors’ hospital, and his drowsy state interfered with detailed interviews. According to the chest CT scan performed immediately after admission, coronary stents had been inserted at the bifurcation site of the left main coronary artery (Figure 2). It also appeared that the inserted stents were drug-eluting stents (DES) considering that he had received dual antiplatelet therapy (DAPT) with aspirin and clopidogrel (Plavix®; Sanofi, Bridgewater, NJ, USA) a switching platelet P2Y12 inhibitor, for approximately 4 months. However, DAPT was discontinued owing to the worsening cerebral hemorrhage after hospitalization; bridging antiplatelet therapy was also not performed. Before the surgery, the patient showed a normal electrocardiogram (ECG), and cardiac biomarker testing revealed a troponin-T concentration of 0.006 µg/L, which was within normal limits (Figure 3). The patient demonstrated satisfactory functional capacity to be able to work at a construction site just before the trauma. Therefore, an additional preoperative cardiac evaluation was not considered necessary.
Figure 2. Chest computed tomography (CT) images showing coronary stents in the main bifurcation of the left coronary arteries (white arrow in (a) and (b)).
Figure 3. Preoperative electrocardiogram of the patient showing normal sinus rhythm with a heart rate of 95 bpm.
bpm, beats per minute.
Immediately after admission to the ICU, the patient’s bladder temperature was 38.0°C. Using a hydrogel-coated water-circulating pad (Arctic Sun®; Medivance Inc., Louisville, CO, USA), which is a surface cooling device, the patient underwent mild therapeutic hypothermia with a target temperature of 33.0°C. During therapeutic hypothermia, sedation and shivering control were performed using continuous rate infusions of midazolam at 5 mg/hour and remifentanil at 0.2 µg/kg/minute. The patient’s bladder temperature was 33.4°C just before he was transferred to the operating room. Preoperative laboratory tests showed no unusual findings except for mildly elevated aspartate aminotransferase (73 U/L) and alanine aminotransferase (51 U/L) concentrations.
The patient’s vital signs upon arrival at the operating room were a BP of 130/70 mmHg and SpO2 of 100%. Lead II of the ECG was monitored, and there were no unusual findings except sinus tachycardia with an HR of 100 bpm on the initial ECG. Arterial BP (ABP), central venous pressure (CVP), and esophageal temperature were monitored from the right radial artery, right subclavian vein, and lower third of the esophagus using a temperature sensor attached to the esophageal stethoscope, respectively. The initial ABP, CVP, and esophageal temperature were 140/85 mmHg, 6 mmHg, and 33.0°C, respectively. Arterial blood gas analysis was performed in the arterial line, with values calculated as for a body temperature of 37°C, which revealed the following: pH, 7.400; partial pressure of carbon dioxide (PCO2), 40 mmHg; partial pressure of oxygen (PO2), 174 mmHg; bicarbonate (HCO3), 24.8 mmol/L; base excess, 0.0; lactate, 0.17 mmol/L; ionized calcium, 1.04 mmol/L; and potassium, 3.3 mmol/L. Total intravenous (IV) anesthesia was performed using a target-controlled infusion of propofol and remifentanil in accordance with the Schnider6 and Minto7 models, respectively. Propofol and remifentanil were infused continuously with effect-site concentrations (Ce) of 1.5 µg/mL and 1.5 ng/mL, respectively. Concurrently, 50 mg of rocuronium was administered by IV bolus injection for neuromuscular blockade. Mechanical ventilation was adjusted to maintain the end-tidal carbon dioxide value at 30 to 33 mmHg.
As ABP decreased to 78/42 mmHg and CVP decreased to 4 mmHg within 10 minutes after anesthesia, the Ce of propofol and remifentanil were lowered to 0.8 µg/mL and 1.0 ng/mL, respectively, and approximately 1 L of balanced crystalloid fluid loading was performed. However, the patient’s BP did not recover, and a dopamine continuous infusion (5 µg/kg/minute) was initiated. When a surgical incision was made 10 minutes after beginning the dopamine infusion, the patient’s ABP and HR increased to 115/70 mmHg and 120 bpm, respectively, and marked ST-segment depression (−2.4 mm) was found, which was not observed on the preoperative or initial ECGs (Figure 4). Subsequently, the depth of anesthesia was increased by increasing the Ce of propofol to 1.0 to 1.2 µg/mL, followed by a continuous infusion of isosorbide dinitrate (1 µg/kg/minute). While HR was maintained at 110 to 120 bpm, ABP gradually decreased to 70/40 mmHg after approximately 5 minutes. To treat the hypotension, isosorbide dinitrate was discontinued, and a bolus injection of 100 µg phenylephrine was administered, after which, ABP increased temporarily to 95/60 mmHg. Subsequently, HR decreased suddenly and rapidly, and ventricular fibrillation occurred. The anesthesiologists immediately discontinued the surgery and performed cardiopulmonary resuscitation (CPR). The patient’s esophageal temperature at the time was 32.0°C.
Figure 4. Intraoperative electrocardiogram of the patient showing sinus tachycardia with significant ST depression. The ST depression on the monitor represented approximately −2.4 at the time.
Chest compressions were performed with periodic defibrillation with 200 J, and epinephrine injections were administered every 3 minutes; these measures elicited no response. During CPR, ECG monitoring showed temporary severe bradycardia (HR, 10–30 bpm). However, this was considered pulseless electrical activity because no pulse was detected in the patient’s femoral artery. Soon afterward, there was a rapid transition to ventricular fibrillation, and CPR was repeated. At the time, the patient’s esophageal temperature dropped to 30.0°C. According to the refractory state after the defibrillation and drug injections, we considered that the cardiac arrest was caused by hypothermia. To reverse the cause, a forced-air warming device and a warming blanket (Bair Hugger®; Arizant Healthcare Inc., Eden Prairie, MN, USA) were used for surface warming, the IV fluids were warmed, and surgical wound closure and dressing were performed rapidly to prevent surface heat loss. The first arterial blood gas analysis after ventricular fibrillation, calculated as for a body temperature of 37°C, showed the following: pH, 7.690; PCO2, 15 mmHg; PO2, 260 mmHg; HCO3, 18.1 mmol/L; base excess, −1.2; lactate, 0.33 mmol/L; ionized calcium, 0.88 mmol/L; and potassium, 5.0 mmol/L. Follow-up arterial blood gas analyses were repeated periodically. As we continued to perform CPR, the patient developed severe metabolic acidosis and hypocalcemia characterized by a pH of 6.840, HCO3 of 15.4 mmol/L, and ionized calcium of 0.66 mmol/L. To correct the abnormal findings, we administered appropriate amounts of sodium bicarbonate and calcium chloride. After an hour-long CPR and rewarming, the patient’s esophageal temperature recovered to 33.0°C, which was followed by observing sinus rhythm characterized by an HR of 75 bpm on an ECG monitor and detecting a pulse in the femoral artery with a characteristic ABP waveform that eventually led to the recovery of BP to 130/55 mmHg. Upon confirming recovery of spontaneous circulation (ROSC), the patient was transported to the ICU while we administered dopamine (10 µg/kg/minute), norepinephrine (0.1 µg/kg/minute), and vasopressin (2 U/hour). ABP, HR, and esophageal temperature during anesthesia are shown in Figure 5.
Figure 5. Changes in vital signs (blood pressure (BP), heart rate (HR), and esophageal temperature) over time (5-minute intervals).
bpm, beats per minute.
Cardiac biomarker testing performed after the patient’s transfer to the ICU showed an increased troponin-T level of 1.432 µg/L. Portable Doppler transthoracic echocardiography (TTE) revealed total occlusion of the left main coronary artery, which had undergone PCI previously. Because emergency coronary artery bypass grafting (CABG) was impossible at the time, repeated PCI was considered an alternative treatment option. However, even after his transfer to the ICU, the patient experienced repetitive ventricular fibrillation. Four hours after the patient’s transfer to the ICU, the patient’s guardian requested that CPR should be stopped. Without resuscitation, the patient died.
Discussion
In this case, the patient, who was receiving therapeutic hypothermia, underwent emergency craniectomy owing to worsening brain hemorrhage, and cardiac arrest occurred during general anesthesia. We deduced that the cardiac arrest was owing to the additional decrease in the patient’s core body temperature and myocardial infarction caused by restenosis of the previous PCI-treated lesion.
Therapeutic hypothermia is used to improve neurological outcomes in cardiac arrest, TBI, stroke, and hypoxic–ischemic encephalopathy.1,4 Studies on the efficacy of therapeutic hypothermia for TBI have been less convincing, and they show conflicting results. However, hypothermia is consistently used as an option for TBI because it reduces intracranial pressure and the cerebral metabolic rate, thereby protecting neuronal function in the brain and decreasing the wide range of injuries or reparative processes by attenuating temperature-sensitive mechanisms, such as excitotoxicity, free radical generation, apoptosis, and inflammation.1,2
The cardiovascular effects associated with hypothermia are complex. Hypothermia affects the electrical conduction of myocardial cells, which can induce hypotension or bradycardia and even increase the risk of ventricular fibrillation by prolonging repolarization.4,8 However, mild hypothermia of 32.0°C to 35.0°C demonstrates cardioprotective effects by decreasing myocardial oxygen demand and increasing myocardial perfusion through vasodilation of coronary vessels.9 Therefore, there has been a growing interest in the clinical use of mild therapeutic hypothermia as an adjunctive therapy during ischemic heart disease.10 Our patient received mild therapeutic hypothermia with a target core temperature of 33.0°C, which may have yielded cardioprotective as well as neuroprotective benefits.11
During general anesthesia, hypothermia is induced following an increase in heat loss caused by peripheral vasodilation after the injection of anesthetic drugs, heat redistribution from the core to the periphery, and impaired normal thermoregulatory responses, such as vasoconstriction or shivering.5 In our patient, the cold operating room temperature and irrigating an uncovered surgical area in the brain with cold saline also may have contributed to the additional core body temperature drop in addition to the effects of general anesthesia.12 However, in our case, neurosurgeons requested that anesthesiologists not perform warming during anesthesia. Instead, esophageal temperature and signs that could result from hypothermia, such as ECG changes, were carefully monitored. The esophageal temperature decreased from 33.0°C, which was measured immediately after entering the operation room, to 32.0°C at the time of cardiac arrest. The esophageal temperature further decreased to a minimum of 30.0°C during CPR. With rewarming, the temperature recovered to 33.0°C and finally, ROSC was achieved. Considering that chest compressions, drugs, and defibrillation did not induce ROSC, which was achieved with the recovery of an esophageal temperature of 33.0°C, the additional rapid fall in the patient’s core body temperature during anesthesia may have caused the cardiac arrest.
In this case study, we suspected that restenosis at the site of the previous stent insertion contributed to the cardiac arrest, given the following three findings: First, the patient had a normal preoperative ECG with no signs of ischemia but developed ST-segment depression during anesthesia. Second, repetitive cardiac arrests occurred even after achieving ROSC with rewarming. Third, an increased troponin-T concentration was observed, and suspected re-stenotic occlusion of the left main coronary artery was confirmed by Doppler TTE after the patient’s transfer to the ICU. Accordingly, myocardial infarction was considered the main cause of death.
The mechanisms of myocardial infarction owing to stent restenosis after DES insertion are rapid stent thrombosis or profuse in-stent restenosis (ISR).13 We assume that the lesion characteristics in the left main bifurcation of the coronary artery and DAPT interruption owing to the increasing amount of cerebral hemorrhage contributed to the stent restenosis in this patient. As evidence in support of this assumption, previous studies have shown that stenting at the left main bifurcation and DAPT interruption are strong risk factors for stent thrombosis or ISR.14,15 While controversial, not interrupting aspirin perioperatively could have avoided poor outcomes.16 However, there were no clinical signs related to expected ongoing restenosis of the coronary artery or myocardial ischemia preoperatively in our patient. We suspect that the possible ischemic signs may have been masked by the aforementioned cardioprotective benefits of therapeutic hypothermia. We also believe that the following three conditions contributed to the intraoperative myocardial ischemia and infarction caused by acute restenosis of the coronary stent: First, intraoperative hypotension may have contributed to myocardial injury. The IV anesthetics used during anesthesia, propofol and remifentanil, induced redistribution of blood volume by peripheral vasodilation, resulting in profound hypotension.4 Profound hypotension was strongly associated with myocardial injury and especially harmful for this patient, who had preexisting ischemic heart disease.17 Second, tachycardia, which may have been caused by the inotrope dopamine, used to compensate for the low BP or the stress response to surgical stimulation from shallow anesthetic depth, may have triggered myocardial ischemia by decreasing oxygen delivery and increasing oxygen consumption.17,18 If cardiogenic shock is suspected, as in our case, norepinephrine could be a better option than dopamine.19 Third, the bolus injection of phenylephrine, a selective α1-adrenergic agonist, just before cardiac arrest may have temporarily and abruptly increased cardiac afterload and impaired myocardial perfusion.20
Had we known that coronary artery occlusion would result from hypothermia-induced cardiac arrest and coronary stent restenosis, the treatment of choice would have been CABG surgery while maintaining circulation with emergency cardiopulmonary bypass (CPB) or extracorporeal membrane oxygenation (ECMO) via the femoral artery and vein.11 However, it was difficult to determine the etiology of the cardiac arrest at the time. Even if we had known the cause, ideal treatment would have been impossible, given circumstances beyond our control; all thoracic surgeons were busy operating, and emergency CPB and ECMO were not available. As an alternative, repeat PCI after the patient achieved ROSC could have been performed. However, the patient had a prolonged CPR duration, and he did not maintain the ROSC state for a period sufficient for intervention, which led to a poor outcome.
In this case, it was not possible to accurately determine the reason for the cardiac arrest and for not achieving consistent ROSC. Instead, we conclude that cardiac arrest developed and persisted as a result of the unintentional severe hypothermia that occurred during general anesthesia and/or coronary artery restenosis at the site of the previous PCI. A thorough evaluation of heart-related history is required for TBI patients who have a history of ischemic heart diseases and who undergo therapeutic hypothermia because there may be undiscovered coronary insufficiency owing to the cardioprotective effects of therapeutic hypothermia. In addition, careful monitoring of the patient’s core body temperature, and preventive and countermeasures should be fully prepared and taken against adverse hemodynamic effects, even cardiac arrest, which can occur because of additive hypothermic responses during anesthesia.
Ethics statement: This case report was approved by our Institutional Review Board (EMC 2020-05-005-001), and the case report was performed in compliance with the EQUATOR Network guidelines (the CARE guidelines). All patient details were de-identified, and written consent was obtained from the patient’s guardian for treatment and publication of this report.
Declaration of conflicting interest: The authors declare that there is no conflict of interest.
Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
ORCID iDs: Dong Ho Park https://orcid.org/0000-0002-6587-3756
Tae Woo Kim https://orcid.org/0000-0002-2115-6448
Mo Se Kim https://orcid.org/0000-0001-7362-0030 | Intravenous bolus | DrugAdministrationRoute | CC BY-NC | 33499678 | 19,429,456 | 2021-01 |
What was the administration route of drug 'REMIFENTANIL HYDROCHLORIDE'? | Cardiac arrest caused by accidental severe hypothermia and myocardial infarction during general anesthesia.
Therapeutic hypothermia is often used for traumatic brain injury because of its neuroprotective effect and decreased secondary brain injury. However, this procedure lacks clinical evidence supporting its efficacy, and adverse outcomes have been reported during general anesthesia. A 61-year-old man with a history of percutaneous coronary intervention (PCI) was admitted with traumatic brain injury. Immediately after admission, he underwent mild therapeutic hypothermia with a target temperature of 33.0°C for neuroprotection. During general anesthesia for emergency surgery because he developed a mass effect, hypothermic cardiac arrest occurred following an additional decrease in the core body temperature. Moreover, myocardial infarction caused by restenosis of the previous PCI lesion also contributed to the cardiac arrest. Although the patient recovered spontaneous circulation after an hour-long cardiopulmonary resuscitation with rewarming, he eventually died of subsequent repetitive cardiac arrests. When anesthetizing patients undergoing therapeutic hypothermia, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.
Introduction
Therapeutic hypothermia is an attractive strategy for traumatic brain injury (TBI), which demonstrates neuroprotective effects and decreases secondary brain injury.1 Despite the lack of clinical evidence regarding efficacy, therapeutic hypothermia after TBI has been used for at least half a century because of its potential benefits, which are supported by numerous preclinical studies.2,3 However, there have been no reports on the danger of unintentional severe hypothermia during general anesthesia caused by the additional drop in temperature of patients undergoing therapeutic hypothermia. When anesthetizing patients with comorbidities, such as ischemic heart disease, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.4,5
Case report
A 61-year-old man (height, 168 cm; weight, 70 kg) presented to our emergency room (ER) with head trauma after falling on a cement floor from a 5-meter-tall construction site. Upon arrival at the ER, the patient had a Glasgow coma scale score of 11 and the following vital signs: blood pressure (BP), 147/83 mmHg; heart rate (HR), 86 beats per minute (bpm); respiratory rate, 24 breaths per minute; body temperature from the auditory canal, 36.8°C; and percutaneous oxygen saturation (SpO2), 97%. Upon discovering signs of acute subdural hemorrhage, subarachnoid hemorrhage, and intraventricular hemorrhage from a brain computed tomography (CT) scan performed in the ER, we admitted the patient to the intensive care unit (ICU) for close monitoring. After approximately 6 hours, we performed follow-up brain CT because the patient’s mental status had deteriorated from drowsy to stupor. Upon observing a mass effect characterized by an increase in the hemorrhage volume and a left midline shift on the CT images, the patient was scheduled for immediate intubation and emergency decompressive craniectomy (Figure 1).
Figure 1. Brain computed tomography (CT) images taken 6 hours after intensive care unit admission. (a) Axial view showing acute subarachnoid hemorrhage in the peri-mesencephalic cistern (arrow A) and subdural hemorrhage in the right inferior temporal (arrow B) and right parietal (arrow C) regions. (b) Axial view showing acute subdural hemorrhage in the right temporal region (black arrow) and left midline shift.
Although the patient had a history of ischemic heart disease with percutaneous coronary intervention (PCI), it was difficult to obtain accurate information because there was no relevant data in the patient’s medical records from the authors’ hospital, and his drowsy state interfered with detailed interviews. According to the chest CT scan performed immediately after admission, coronary stents had been inserted at the bifurcation site of the left main coronary artery (Figure 2). It also appeared that the inserted stents were drug-eluting stents (DES) considering that he had received dual antiplatelet therapy (DAPT) with aspirin and clopidogrel (Plavix®; Sanofi, Bridgewater, NJ, USA) a switching platelet P2Y12 inhibitor, for approximately 4 months. However, DAPT was discontinued owing to the worsening cerebral hemorrhage after hospitalization; bridging antiplatelet therapy was also not performed. Before the surgery, the patient showed a normal electrocardiogram (ECG), and cardiac biomarker testing revealed a troponin-T concentration of 0.006 µg/L, which was within normal limits (Figure 3). The patient demonstrated satisfactory functional capacity to be able to work at a construction site just before the trauma. Therefore, an additional preoperative cardiac evaluation was not considered necessary.
Figure 2. Chest computed tomography (CT) images showing coronary stents in the main bifurcation of the left coronary arteries (white arrow in (a) and (b)).
Figure 3. Preoperative electrocardiogram of the patient showing normal sinus rhythm with a heart rate of 95 bpm.
bpm, beats per minute.
Immediately after admission to the ICU, the patient’s bladder temperature was 38.0°C. Using a hydrogel-coated water-circulating pad (Arctic Sun®; Medivance Inc., Louisville, CO, USA), which is a surface cooling device, the patient underwent mild therapeutic hypothermia with a target temperature of 33.0°C. During therapeutic hypothermia, sedation and shivering control were performed using continuous rate infusions of midazolam at 5 mg/hour and remifentanil at 0.2 µg/kg/minute. The patient’s bladder temperature was 33.4°C just before he was transferred to the operating room. Preoperative laboratory tests showed no unusual findings except for mildly elevated aspartate aminotransferase (73 U/L) and alanine aminotransferase (51 U/L) concentrations.
The patient’s vital signs upon arrival at the operating room were a BP of 130/70 mmHg and SpO2 of 100%. Lead II of the ECG was monitored, and there were no unusual findings except sinus tachycardia with an HR of 100 bpm on the initial ECG. Arterial BP (ABP), central venous pressure (CVP), and esophageal temperature were monitored from the right radial artery, right subclavian vein, and lower third of the esophagus using a temperature sensor attached to the esophageal stethoscope, respectively. The initial ABP, CVP, and esophageal temperature were 140/85 mmHg, 6 mmHg, and 33.0°C, respectively. Arterial blood gas analysis was performed in the arterial line, with values calculated as for a body temperature of 37°C, which revealed the following: pH, 7.400; partial pressure of carbon dioxide (PCO2), 40 mmHg; partial pressure of oxygen (PO2), 174 mmHg; bicarbonate (HCO3), 24.8 mmol/L; base excess, 0.0; lactate, 0.17 mmol/L; ionized calcium, 1.04 mmol/L; and potassium, 3.3 mmol/L. Total intravenous (IV) anesthesia was performed using a target-controlled infusion of propofol and remifentanil in accordance with the Schnider6 and Minto7 models, respectively. Propofol and remifentanil were infused continuously with effect-site concentrations (Ce) of 1.5 µg/mL and 1.5 ng/mL, respectively. Concurrently, 50 mg of rocuronium was administered by IV bolus injection for neuromuscular blockade. Mechanical ventilation was adjusted to maintain the end-tidal carbon dioxide value at 30 to 33 mmHg.
As ABP decreased to 78/42 mmHg and CVP decreased to 4 mmHg within 10 minutes after anesthesia, the Ce of propofol and remifentanil were lowered to 0.8 µg/mL and 1.0 ng/mL, respectively, and approximately 1 L of balanced crystalloid fluid loading was performed. However, the patient’s BP did not recover, and a dopamine continuous infusion (5 µg/kg/minute) was initiated. When a surgical incision was made 10 minutes after beginning the dopamine infusion, the patient’s ABP and HR increased to 115/70 mmHg and 120 bpm, respectively, and marked ST-segment depression (−2.4 mm) was found, which was not observed on the preoperative or initial ECGs (Figure 4). Subsequently, the depth of anesthesia was increased by increasing the Ce of propofol to 1.0 to 1.2 µg/mL, followed by a continuous infusion of isosorbide dinitrate (1 µg/kg/minute). While HR was maintained at 110 to 120 bpm, ABP gradually decreased to 70/40 mmHg after approximately 5 minutes. To treat the hypotension, isosorbide dinitrate was discontinued, and a bolus injection of 100 µg phenylephrine was administered, after which, ABP increased temporarily to 95/60 mmHg. Subsequently, HR decreased suddenly and rapidly, and ventricular fibrillation occurred. The anesthesiologists immediately discontinued the surgery and performed cardiopulmonary resuscitation (CPR). The patient’s esophageal temperature at the time was 32.0°C.
Figure 4. Intraoperative electrocardiogram of the patient showing sinus tachycardia with significant ST depression. The ST depression on the monitor represented approximately −2.4 at the time.
Chest compressions were performed with periodic defibrillation with 200 J, and epinephrine injections were administered every 3 minutes; these measures elicited no response. During CPR, ECG monitoring showed temporary severe bradycardia (HR, 10–30 bpm). However, this was considered pulseless electrical activity because no pulse was detected in the patient’s femoral artery. Soon afterward, there was a rapid transition to ventricular fibrillation, and CPR was repeated. At the time, the patient’s esophageal temperature dropped to 30.0°C. According to the refractory state after the defibrillation and drug injections, we considered that the cardiac arrest was caused by hypothermia. To reverse the cause, a forced-air warming device and a warming blanket (Bair Hugger®; Arizant Healthcare Inc., Eden Prairie, MN, USA) were used for surface warming, the IV fluids were warmed, and surgical wound closure and dressing were performed rapidly to prevent surface heat loss. The first arterial blood gas analysis after ventricular fibrillation, calculated as for a body temperature of 37°C, showed the following: pH, 7.690; PCO2, 15 mmHg; PO2, 260 mmHg; HCO3, 18.1 mmol/L; base excess, −1.2; lactate, 0.33 mmol/L; ionized calcium, 0.88 mmol/L; and potassium, 5.0 mmol/L. Follow-up arterial blood gas analyses were repeated periodically. As we continued to perform CPR, the patient developed severe metabolic acidosis and hypocalcemia characterized by a pH of 6.840, HCO3 of 15.4 mmol/L, and ionized calcium of 0.66 mmol/L. To correct the abnormal findings, we administered appropriate amounts of sodium bicarbonate and calcium chloride. After an hour-long CPR and rewarming, the patient’s esophageal temperature recovered to 33.0°C, which was followed by observing sinus rhythm characterized by an HR of 75 bpm on an ECG monitor and detecting a pulse in the femoral artery with a characteristic ABP waveform that eventually led to the recovery of BP to 130/55 mmHg. Upon confirming recovery of spontaneous circulation (ROSC), the patient was transported to the ICU while we administered dopamine (10 µg/kg/minute), norepinephrine (0.1 µg/kg/minute), and vasopressin (2 U/hour). ABP, HR, and esophageal temperature during anesthesia are shown in Figure 5.
Figure 5. Changes in vital signs (blood pressure (BP), heart rate (HR), and esophageal temperature) over time (5-minute intervals).
bpm, beats per minute.
Cardiac biomarker testing performed after the patient’s transfer to the ICU showed an increased troponin-T level of 1.432 µg/L. Portable Doppler transthoracic echocardiography (TTE) revealed total occlusion of the left main coronary artery, which had undergone PCI previously. Because emergency coronary artery bypass grafting (CABG) was impossible at the time, repeated PCI was considered an alternative treatment option. However, even after his transfer to the ICU, the patient experienced repetitive ventricular fibrillation. Four hours after the patient’s transfer to the ICU, the patient’s guardian requested that CPR should be stopped. Without resuscitation, the patient died.
Discussion
In this case, the patient, who was receiving therapeutic hypothermia, underwent emergency craniectomy owing to worsening brain hemorrhage, and cardiac arrest occurred during general anesthesia. We deduced that the cardiac arrest was owing to the additional decrease in the patient’s core body temperature and myocardial infarction caused by restenosis of the previous PCI-treated lesion.
Therapeutic hypothermia is used to improve neurological outcomes in cardiac arrest, TBI, stroke, and hypoxic–ischemic encephalopathy.1,4 Studies on the efficacy of therapeutic hypothermia for TBI have been less convincing, and they show conflicting results. However, hypothermia is consistently used as an option for TBI because it reduces intracranial pressure and the cerebral metabolic rate, thereby protecting neuronal function in the brain and decreasing the wide range of injuries or reparative processes by attenuating temperature-sensitive mechanisms, such as excitotoxicity, free radical generation, apoptosis, and inflammation.1,2
The cardiovascular effects associated with hypothermia are complex. Hypothermia affects the electrical conduction of myocardial cells, which can induce hypotension or bradycardia and even increase the risk of ventricular fibrillation by prolonging repolarization.4,8 However, mild hypothermia of 32.0°C to 35.0°C demonstrates cardioprotective effects by decreasing myocardial oxygen demand and increasing myocardial perfusion through vasodilation of coronary vessels.9 Therefore, there has been a growing interest in the clinical use of mild therapeutic hypothermia as an adjunctive therapy during ischemic heart disease.10 Our patient received mild therapeutic hypothermia with a target core temperature of 33.0°C, which may have yielded cardioprotective as well as neuroprotective benefits.11
During general anesthesia, hypothermia is induced following an increase in heat loss caused by peripheral vasodilation after the injection of anesthetic drugs, heat redistribution from the core to the periphery, and impaired normal thermoregulatory responses, such as vasoconstriction or shivering.5 In our patient, the cold operating room temperature and irrigating an uncovered surgical area in the brain with cold saline also may have contributed to the additional core body temperature drop in addition to the effects of general anesthesia.12 However, in our case, neurosurgeons requested that anesthesiologists not perform warming during anesthesia. Instead, esophageal temperature and signs that could result from hypothermia, such as ECG changes, were carefully monitored. The esophageal temperature decreased from 33.0°C, which was measured immediately after entering the operation room, to 32.0°C at the time of cardiac arrest. The esophageal temperature further decreased to a minimum of 30.0°C during CPR. With rewarming, the temperature recovered to 33.0°C and finally, ROSC was achieved. Considering that chest compressions, drugs, and defibrillation did not induce ROSC, which was achieved with the recovery of an esophageal temperature of 33.0°C, the additional rapid fall in the patient’s core body temperature during anesthesia may have caused the cardiac arrest.
In this case study, we suspected that restenosis at the site of the previous stent insertion contributed to the cardiac arrest, given the following three findings: First, the patient had a normal preoperative ECG with no signs of ischemia but developed ST-segment depression during anesthesia. Second, repetitive cardiac arrests occurred even after achieving ROSC with rewarming. Third, an increased troponin-T concentration was observed, and suspected re-stenotic occlusion of the left main coronary artery was confirmed by Doppler TTE after the patient’s transfer to the ICU. Accordingly, myocardial infarction was considered the main cause of death.
The mechanisms of myocardial infarction owing to stent restenosis after DES insertion are rapid stent thrombosis or profuse in-stent restenosis (ISR).13 We assume that the lesion characteristics in the left main bifurcation of the coronary artery and DAPT interruption owing to the increasing amount of cerebral hemorrhage contributed to the stent restenosis in this patient. As evidence in support of this assumption, previous studies have shown that stenting at the left main bifurcation and DAPT interruption are strong risk factors for stent thrombosis or ISR.14,15 While controversial, not interrupting aspirin perioperatively could have avoided poor outcomes.16 However, there were no clinical signs related to expected ongoing restenosis of the coronary artery or myocardial ischemia preoperatively in our patient. We suspect that the possible ischemic signs may have been masked by the aforementioned cardioprotective benefits of therapeutic hypothermia. We also believe that the following three conditions contributed to the intraoperative myocardial ischemia and infarction caused by acute restenosis of the coronary stent: First, intraoperative hypotension may have contributed to myocardial injury. The IV anesthetics used during anesthesia, propofol and remifentanil, induced redistribution of blood volume by peripheral vasodilation, resulting in profound hypotension.4 Profound hypotension was strongly associated with myocardial injury and especially harmful for this patient, who had preexisting ischemic heart disease.17 Second, tachycardia, which may have been caused by the inotrope dopamine, used to compensate for the low BP or the stress response to surgical stimulation from shallow anesthetic depth, may have triggered myocardial ischemia by decreasing oxygen delivery and increasing oxygen consumption.17,18 If cardiogenic shock is suspected, as in our case, norepinephrine could be a better option than dopamine.19 Third, the bolus injection of phenylephrine, a selective α1-adrenergic agonist, just before cardiac arrest may have temporarily and abruptly increased cardiac afterload and impaired myocardial perfusion.20
Had we known that coronary artery occlusion would result from hypothermia-induced cardiac arrest and coronary stent restenosis, the treatment of choice would have been CABG surgery while maintaining circulation with emergency cardiopulmonary bypass (CPB) or extracorporeal membrane oxygenation (ECMO) via the femoral artery and vein.11 However, it was difficult to determine the etiology of the cardiac arrest at the time. Even if we had known the cause, ideal treatment would have been impossible, given circumstances beyond our control; all thoracic surgeons were busy operating, and emergency CPB and ECMO were not available. As an alternative, repeat PCI after the patient achieved ROSC could have been performed. However, the patient had a prolonged CPR duration, and he did not maintain the ROSC state for a period sufficient for intervention, which led to a poor outcome.
In this case, it was not possible to accurately determine the reason for the cardiac arrest and for not achieving consistent ROSC. Instead, we conclude that cardiac arrest developed and persisted as a result of the unintentional severe hypothermia that occurred during general anesthesia and/or coronary artery restenosis at the site of the previous PCI. A thorough evaluation of heart-related history is required for TBI patients who have a history of ischemic heart diseases and who undergo therapeutic hypothermia because there may be undiscovered coronary insufficiency owing to the cardioprotective effects of therapeutic hypothermia. In addition, careful monitoring of the patient’s core body temperature, and preventive and countermeasures should be fully prepared and taken against adverse hemodynamic effects, even cardiac arrest, which can occur because of additive hypothermic responses during anesthesia.
Ethics statement: This case report was approved by our Institutional Review Board (EMC 2020-05-005-001), and the case report was performed in compliance with the EQUATOR Network guidelines (the CARE guidelines). All patient details were de-identified, and written consent was obtained from the patient’s guardian for treatment and publication of this report.
Declaration of conflicting interest: The authors declare that there is no conflict of interest.
Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
ORCID iDs: Dong Ho Park https://orcid.org/0000-0002-6587-3756
Tae Woo Kim https://orcid.org/0000-0002-2115-6448
Mo Se Kim https://orcid.org/0000-0001-7362-0030 | Intravenous drip | DrugAdministrationRoute | CC BY-NC | 33499678 | 19,429,456 | 2021-01 |
What was the administration route of drug 'ROCURONIUM BROMIDE'? | Cardiac arrest caused by accidental severe hypothermia and myocardial infarction during general anesthesia.
Therapeutic hypothermia is often used for traumatic brain injury because of its neuroprotective effect and decreased secondary brain injury. However, this procedure lacks clinical evidence supporting its efficacy, and adverse outcomes have been reported during general anesthesia. A 61-year-old man with a history of percutaneous coronary intervention (PCI) was admitted with traumatic brain injury. Immediately after admission, he underwent mild therapeutic hypothermia with a target temperature of 33.0°C for neuroprotection. During general anesthesia for emergency surgery because he developed a mass effect, hypothermic cardiac arrest occurred following an additional decrease in the core body temperature. Moreover, myocardial infarction caused by restenosis of the previous PCI lesion also contributed to the cardiac arrest. Although the patient recovered spontaneous circulation after an hour-long cardiopulmonary resuscitation with rewarming, he eventually died of subsequent repetitive cardiac arrests. When anesthetizing patients undergoing therapeutic hypothermia, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.
Introduction
Therapeutic hypothermia is an attractive strategy for traumatic brain injury (TBI), which demonstrates neuroprotective effects and decreases secondary brain injury.1 Despite the lack of clinical evidence regarding efficacy, therapeutic hypothermia after TBI has been used for at least half a century because of its potential benefits, which are supported by numerous preclinical studies.2,3 However, there have been no reports on the danger of unintentional severe hypothermia during general anesthesia caused by the additional drop in temperature of patients undergoing therapeutic hypothermia. When anesthetizing patients with comorbidities, such as ischemic heart disease, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.4,5
Case report
A 61-year-old man (height, 168 cm; weight, 70 kg) presented to our emergency room (ER) with head trauma after falling on a cement floor from a 5-meter-tall construction site. Upon arrival at the ER, the patient had a Glasgow coma scale score of 11 and the following vital signs: blood pressure (BP), 147/83 mmHg; heart rate (HR), 86 beats per minute (bpm); respiratory rate, 24 breaths per minute; body temperature from the auditory canal, 36.8°C; and percutaneous oxygen saturation (SpO2), 97%. Upon discovering signs of acute subdural hemorrhage, subarachnoid hemorrhage, and intraventricular hemorrhage from a brain computed tomography (CT) scan performed in the ER, we admitted the patient to the intensive care unit (ICU) for close monitoring. After approximately 6 hours, we performed follow-up brain CT because the patient’s mental status had deteriorated from drowsy to stupor. Upon observing a mass effect characterized by an increase in the hemorrhage volume and a left midline shift on the CT images, the patient was scheduled for immediate intubation and emergency decompressive craniectomy (Figure 1).
Figure 1. Brain computed tomography (CT) images taken 6 hours after intensive care unit admission. (a) Axial view showing acute subarachnoid hemorrhage in the peri-mesencephalic cistern (arrow A) and subdural hemorrhage in the right inferior temporal (arrow B) and right parietal (arrow C) regions. (b) Axial view showing acute subdural hemorrhage in the right temporal region (black arrow) and left midline shift.
Although the patient had a history of ischemic heart disease with percutaneous coronary intervention (PCI), it was difficult to obtain accurate information because there was no relevant data in the patient’s medical records from the authors’ hospital, and his drowsy state interfered with detailed interviews. According to the chest CT scan performed immediately after admission, coronary stents had been inserted at the bifurcation site of the left main coronary artery (Figure 2). It also appeared that the inserted stents were drug-eluting stents (DES) considering that he had received dual antiplatelet therapy (DAPT) with aspirin and clopidogrel (Plavix®; Sanofi, Bridgewater, NJ, USA) a switching platelet P2Y12 inhibitor, for approximately 4 months. However, DAPT was discontinued owing to the worsening cerebral hemorrhage after hospitalization; bridging antiplatelet therapy was also not performed. Before the surgery, the patient showed a normal electrocardiogram (ECG), and cardiac biomarker testing revealed a troponin-T concentration of 0.006 µg/L, which was within normal limits (Figure 3). The patient demonstrated satisfactory functional capacity to be able to work at a construction site just before the trauma. Therefore, an additional preoperative cardiac evaluation was not considered necessary.
Figure 2. Chest computed tomography (CT) images showing coronary stents in the main bifurcation of the left coronary arteries (white arrow in (a) and (b)).
Figure 3. Preoperative electrocardiogram of the patient showing normal sinus rhythm with a heart rate of 95 bpm.
bpm, beats per minute.
Immediately after admission to the ICU, the patient’s bladder temperature was 38.0°C. Using a hydrogel-coated water-circulating pad (Arctic Sun®; Medivance Inc., Louisville, CO, USA), which is a surface cooling device, the patient underwent mild therapeutic hypothermia with a target temperature of 33.0°C. During therapeutic hypothermia, sedation and shivering control were performed using continuous rate infusions of midazolam at 5 mg/hour and remifentanil at 0.2 µg/kg/minute. The patient’s bladder temperature was 33.4°C just before he was transferred to the operating room. Preoperative laboratory tests showed no unusual findings except for mildly elevated aspartate aminotransferase (73 U/L) and alanine aminotransferase (51 U/L) concentrations.
The patient’s vital signs upon arrival at the operating room were a BP of 130/70 mmHg and SpO2 of 100%. Lead II of the ECG was monitored, and there were no unusual findings except sinus tachycardia with an HR of 100 bpm on the initial ECG. Arterial BP (ABP), central venous pressure (CVP), and esophageal temperature were monitored from the right radial artery, right subclavian vein, and lower third of the esophagus using a temperature sensor attached to the esophageal stethoscope, respectively. The initial ABP, CVP, and esophageal temperature were 140/85 mmHg, 6 mmHg, and 33.0°C, respectively. Arterial blood gas analysis was performed in the arterial line, with values calculated as for a body temperature of 37°C, which revealed the following: pH, 7.400; partial pressure of carbon dioxide (PCO2), 40 mmHg; partial pressure of oxygen (PO2), 174 mmHg; bicarbonate (HCO3), 24.8 mmol/L; base excess, 0.0; lactate, 0.17 mmol/L; ionized calcium, 1.04 mmol/L; and potassium, 3.3 mmol/L. Total intravenous (IV) anesthesia was performed using a target-controlled infusion of propofol and remifentanil in accordance with the Schnider6 and Minto7 models, respectively. Propofol and remifentanil were infused continuously with effect-site concentrations (Ce) of 1.5 µg/mL and 1.5 ng/mL, respectively. Concurrently, 50 mg of rocuronium was administered by IV bolus injection for neuromuscular blockade. Mechanical ventilation was adjusted to maintain the end-tidal carbon dioxide value at 30 to 33 mmHg.
As ABP decreased to 78/42 mmHg and CVP decreased to 4 mmHg within 10 minutes after anesthesia, the Ce of propofol and remifentanil were lowered to 0.8 µg/mL and 1.0 ng/mL, respectively, and approximately 1 L of balanced crystalloid fluid loading was performed. However, the patient’s BP did not recover, and a dopamine continuous infusion (5 µg/kg/minute) was initiated. When a surgical incision was made 10 minutes after beginning the dopamine infusion, the patient’s ABP and HR increased to 115/70 mmHg and 120 bpm, respectively, and marked ST-segment depression (−2.4 mm) was found, which was not observed on the preoperative or initial ECGs (Figure 4). Subsequently, the depth of anesthesia was increased by increasing the Ce of propofol to 1.0 to 1.2 µg/mL, followed by a continuous infusion of isosorbide dinitrate (1 µg/kg/minute). While HR was maintained at 110 to 120 bpm, ABP gradually decreased to 70/40 mmHg after approximately 5 minutes. To treat the hypotension, isosorbide dinitrate was discontinued, and a bolus injection of 100 µg phenylephrine was administered, after which, ABP increased temporarily to 95/60 mmHg. Subsequently, HR decreased suddenly and rapidly, and ventricular fibrillation occurred. The anesthesiologists immediately discontinued the surgery and performed cardiopulmonary resuscitation (CPR). The patient’s esophageal temperature at the time was 32.0°C.
Figure 4. Intraoperative electrocardiogram of the patient showing sinus tachycardia with significant ST depression. The ST depression on the monitor represented approximately −2.4 at the time.
Chest compressions were performed with periodic defibrillation with 200 J, and epinephrine injections were administered every 3 minutes; these measures elicited no response. During CPR, ECG monitoring showed temporary severe bradycardia (HR, 10–30 bpm). However, this was considered pulseless electrical activity because no pulse was detected in the patient’s femoral artery. Soon afterward, there was a rapid transition to ventricular fibrillation, and CPR was repeated. At the time, the patient’s esophageal temperature dropped to 30.0°C. According to the refractory state after the defibrillation and drug injections, we considered that the cardiac arrest was caused by hypothermia. To reverse the cause, a forced-air warming device and a warming blanket (Bair Hugger®; Arizant Healthcare Inc., Eden Prairie, MN, USA) were used for surface warming, the IV fluids were warmed, and surgical wound closure and dressing were performed rapidly to prevent surface heat loss. The first arterial blood gas analysis after ventricular fibrillation, calculated as for a body temperature of 37°C, showed the following: pH, 7.690; PCO2, 15 mmHg; PO2, 260 mmHg; HCO3, 18.1 mmol/L; base excess, −1.2; lactate, 0.33 mmol/L; ionized calcium, 0.88 mmol/L; and potassium, 5.0 mmol/L. Follow-up arterial blood gas analyses were repeated periodically. As we continued to perform CPR, the patient developed severe metabolic acidosis and hypocalcemia characterized by a pH of 6.840, HCO3 of 15.4 mmol/L, and ionized calcium of 0.66 mmol/L. To correct the abnormal findings, we administered appropriate amounts of sodium bicarbonate and calcium chloride. After an hour-long CPR and rewarming, the patient’s esophageal temperature recovered to 33.0°C, which was followed by observing sinus rhythm characterized by an HR of 75 bpm on an ECG monitor and detecting a pulse in the femoral artery with a characteristic ABP waveform that eventually led to the recovery of BP to 130/55 mmHg. Upon confirming recovery of spontaneous circulation (ROSC), the patient was transported to the ICU while we administered dopamine (10 µg/kg/minute), norepinephrine (0.1 µg/kg/minute), and vasopressin (2 U/hour). ABP, HR, and esophageal temperature during anesthesia are shown in Figure 5.
Figure 5. Changes in vital signs (blood pressure (BP), heart rate (HR), and esophageal temperature) over time (5-minute intervals).
bpm, beats per minute.
Cardiac biomarker testing performed after the patient’s transfer to the ICU showed an increased troponin-T level of 1.432 µg/L. Portable Doppler transthoracic echocardiography (TTE) revealed total occlusion of the left main coronary artery, which had undergone PCI previously. Because emergency coronary artery bypass grafting (CABG) was impossible at the time, repeated PCI was considered an alternative treatment option. However, even after his transfer to the ICU, the patient experienced repetitive ventricular fibrillation. Four hours after the patient’s transfer to the ICU, the patient’s guardian requested that CPR should be stopped. Without resuscitation, the patient died.
Discussion
In this case, the patient, who was receiving therapeutic hypothermia, underwent emergency craniectomy owing to worsening brain hemorrhage, and cardiac arrest occurred during general anesthesia. We deduced that the cardiac arrest was owing to the additional decrease in the patient’s core body temperature and myocardial infarction caused by restenosis of the previous PCI-treated lesion.
Therapeutic hypothermia is used to improve neurological outcomes in cardiac arrest, TBI, stroke, and hypoxic–ischemic encephalopathy.1,4 Studies on the efficacy of therapeutic hypothermia for TBI have been less convincing, and they show conflicting results. However, hypothermia is consistently used as an option for TBI because it reduces intracranial pressure and the cerebral metabolic rate, thereby protecting neuronal function in the brain and decreasing the wide range of injuries or reparative processes by attenuating temperature-sensitive mechanisms, such as excitotoxicity, free radical generation, apoptosis, and inflammation.1,2
The cardiovascular effects associated with hypothermia are complex. Hypothermia affects the electrical conduction of myocardial cells, which can induce hypotension or bradycardia and even increase the risk of ventricular fibrillation by prolonging repolarization.4,8 However, mild hypothermia of 32.0°C to 35.0°C demonstrates cardioprotective effects by decreasing myocardial oxygen demand and increasing myocardial perfusion through vasodilation of coronary vessels.9 Therefore, there has been a growing interest in the clinical use of mild therapeutic hypothermia as an adjunctive therapy during ischemic heart disease.10 Our patient received mild therapeutic hypothermia with a target core temperature of 33.0°C, which may have yielded cardioprotective as well as neuroprotective benefits.11
During general anesthesia, hypothermia is induced following an increase in heat loss caused by peripheral vasodilation after the injection of anesthetic drugs, heat redistribution from the core to the periphery, and impaired normal thermoregulatory responses, such as vasoconstriction or shivering.5 In our patient, the cold operating room temperature and irrigating an uncovered surgical area in the brain with cold saline also may have contributed to the additional core body temperature drop in addition to the effects of general anesthesia.12 However, in our case, neurosurgeons requested that anesthesiologists not perform warming during anesthesia. Instead, esophageal temperature and signs that could result from hypothermia, such as ECG changes, were carefully monitored. The esophageal temperature decreased from 33.0°C, which was measured immediately after entering the operation room, to 32.0°C at the time of cardiac arrest. The esophageal temperature further decreased to a minimum of 30.0°C during CPR. With rewarming, the temperature recovered to 33.0°C and finally, ROSC was achieved. Considering that chest compressions, drugs, and defibrillation did not induce ROSC, which was achieved with the recovery of an esophageal temperature of 33.0°C, the additional rapid fall in the patient’s core body temperature during anesthesia may have caused the cardiac arrest.
In this case study, we suspected that restenosis at the site of the previous stent insertion contributed to the cardiac arrest, given the following three findings: First, the patient had a normal preoperative ECG with no signs of ischemia but developed ST-segment depression during anesthesia. Second, repetitive cardiac arrests occurred even after achieving ROSC with rewarming. Third, an increased troponin-T concentration was observed, and suspected re-stenotic occlusion of the left main coronary artery was confirmed by Doppler TTE after the patient’s transfer to the ICU. Accordingly, myocardial infarction was considered the main cause of death.
The mechanisms of myocardial infarction owing to stent restenosis after DES insertion are rapid stent thrombosis or profuse in-stent restenosis (ISR).13 We assume that the lesion characteristics in the left main bifurcation of the coronary artery and DAPT interruption owing to the increasing amount of cerebral hemorrhage contributed to the stent restenosis in this patient. As evidence in support of this assumption, previous studies have shown that stenting at the left main bifurcation and DAPT interruption are strong risk factors for stent thrombosis or ISR.14,15 While controversial, not interrupting aspirin perioperatively could have avoided poor outcomes.16 However, there were no clinical signs related to expected ongoing restenosis of the coronary artery or myocardial ischemia preoperatively in our patient. We suspect that the possible ischemic signs may have been masked by the aforementioned cardioprotective benefits of therapeutic hypothermia. We also believe that the following three conditions contributed to the intraoperative myocardial ischemia and infarction caused by acute restenosis of the coronary stent: First, intraoperative hypotension may have contributed to myocardial injury. The IV anesthetics used during anesthesia, propofol and remifentanil, induced redistribution of blood volume by peripheral vasodilation, resulting in profound hypotension.4 Profound hypotension was strongly associated with myocardial injury and especially harmful for this patient, who had preexisting ischemic heart disease.17 Second, tachycardia, which may have been caused by the inotrope dopamine, used to compensate for the low BP or the stress response to surgical stimulation from shallow anesthetic depth, may have triggered myocardial ischemia by decreasing oxygen delivery and increasing oxygen consumption.17,18 If cardiogenic shock is suspected, as in our case, norepinephrine could be a better option than dopamine.19 Third, the bolus injection of phenylephrine, a selective α1-adrenergic agonist, just before cardiac arrest may have temporarily and abruptly increased cardiac afterload and impaired myocardial perfusion.20
Had we known that coronary artery occlusion would result from hypothermia-induced cardiac arrest and coronary stent restenosis, the treatment of choice would have been CABG surgery while maintaining circulation with emergency cardiopulmonary bypass (CPB) or extracorporeal membrane oxygenation (ECMO) via the femoral artery and vein.11 However, it was difficult to determine the etiology of the cardiac arrest at the time. Even if we had known the cause, ideal treatment would have been impossible, given circumstances beyond our control; all thoracic surgeons were busy operating, and emergency CPB and ECMO were not available. As an alternative, repeat PCI after the patient achieved ROSC could have been performed. However, the patient had a prolonged CPR duration, and he did not maintain the ROSC state for a period sufficient for intervention, which led to a poor outcome.
In this case, it was not possible to accurately determine the reason for the cardiac arrest and for not achieving consistent ROSC. Instead, we conclude that cardiac arrest developed and persisted as a result of the unintentional severe hypothermia that occurred during general anesthesia and/or coronary artery restenosis at the site of the previous PCI. A thorough evaluation of heart-related history is required for TBI patients who have a history of ischemic heart diseases and who undergo therapeutic hypothermia because there may be undiscovered coronary insufficiency owing to the cardioprotective effects of therapeutic hypothermia. In addition, careful monitoring of the patient’s core body temperature, and preventive and countermeasures should be fully prepared and taken against adverse hemodynamic effects, even cardiac arrest, which can occur because of additive hypothermic responses during anesthesia.
Ethics statement: This case report was approved by our Institutional Review Board (EMC 2020-05-005-001), and the case report was performed in compliance with the EQUATOR Network guidelines (the CARE guidelines). All patient details were de-identified, and written consent was obtained from the patient’s guardian for treatment and publication of this report.
Declaration of conflicting interest: The authors declare that there is no conflict of interest.
Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
ORCID iDs: Dong Ho Park https://orcid.org/0000-0002-6587-3756
Tae Woo Kim https://orcid.org/0000-0002-2115-6448
Mo Se Kim https://orcid.org/0000-0001-7362-0030 | Intravenous bolus | DrugAdministrationRoute | CC BY-NC | 33499678 | 19,429,456 | 2021-01 |
What was the dosage of drug 'PHENYLEPHRINE HYDROCHLORIDE'? | Cardiac arrest caused by accidental severe hypothermia and myocardial infarction during general anesthesia.
Therapeutic hypothermia is often used for traumatic brain injury because of its neuroprotective effect and decreased secondary brain injury. However, this procedure lacks clinical evidence supporting its efficacy, and adverse outcomes have been reported during general anesthesia. A 61-year-old man with a history of percutaneous coronary intervention (PCI) was admitted with traumatic brain injury. Immediately after admission, he underwent mild therapeutic hypothermia with a target temperature of 33.0°C for neuroprotection. During general anesthesia for emergency surgery because he developed a mass effect, hypothermic cardiac arrest occurred following an additional decrease in the core body temperature. Moreover, myocardial infarction caused by restenosis of the previous PCI lesion also contributed to the cardiac arrest. Although the patient recovered spontaneous circulation after an hour-long cardiopulmonary resuscitation with rewarming, he eventually died of subsequent repetitive cardiac arrests. When anesthetizing patients undergoing therapeutic hypothermia, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.
Introduction
Therapeutic hypothermia is an attractive strategy for traumatic brain injury (TBI), which demonstrates neuroprotective effects and decreases secondary brain injury.1 Despite the lack of clinical evidence regarding efficacy, therapeutic hypothermia after TBI has been used for at least half a century because of its potential benefits, which are supported by numerous preclinical studies.2,3 However, there have been no reports on the danger of unintentional severe hypothermia during general anesthesia caused by the additional drop in temperature of patients undergoing therapeutic hypothermia. When anesthetizing patients with comorbidities, such as ischemic heart disease, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.4,5
Case report
A 61-year-old man (height, 168 cm; weight, 70 kg) presented to our emergency room (ER) with head trauma after falling on a cement floor from a 5-meter-tall construction site. Upon arrival at the ER, the patient had a Glasgow coma scale score of 11 and the following vital signs: blood pressure (BP), 147/83 mmHg; heart rate (HR), 86 beats per minute (bpm); respiratory rate, 24 breaths per minute; body temperature from the auditory canal, 36.8°C; and percutaneous oxygen saturation (SpO2), 97%. Upon discovering signs of acute subdural hemorrhage, subarachnoid hemorrhage, and intraventricular hemorrhage from a brain computed tomography (CT) scan performed in the ER, we admitted the patient to the intensive care unit (ICU) for close monitoring. After approximately 6 hours, we performed follow-up brain CT because the patient’s mental status had deteriorated from drowsy to stupor. Upon observing a mass effect characterized by an increase in the hemorrhage volume and a left midline shift on the CT images, the patient was scheduled for immediate intubation and emergency decompressive craniectomy (Figure 1).
Figure 1. Brain computed tomography (CT) images taken 6 hours after intensive care unit admission. (a) Axial view showing acute subarachnoid hemorrhage in the peri-mesencephalic cistern (arrow A) and subdural hemorrhage in the right inferior temporal (arrow B) and right parietal (arrow C) regions. (b) Axial view showing acute subdural hemorrhage in the right temporal region (black arrow) and left midline shift.
Although the patient had a history of ischemic heart disease with percutaneous coronary intervention (PCI), it was difficult to obtain accurate information because there was no relevant data in the patient’s medical records from the authors’ hospital, and his drowsy state interfered with detailed interviews. According to the chest CT scan performed immediately after admission, coronary stents had been inserted at the bifurcation site of the left main coronary artery (Figure 2). It also appeared that the inserted stents were drug-eluting stents (DES) considering that he had received dual antiplatelet therapy (DAPT) with aspirin and clopidogrel (Plavix®; Sanofi, Bridgewater, NJ, USA) a switching platelet P2Y12 inhibitor, for approximately 4 months. However, DAPT was discontinued owing to the worsening cerebral hemorrhage after hospitalization; bridging antiplatelet therapy was also not performed. Before the surgery, the patient showed a normal electrocardiogram (ECG), and cardiac biomarker testing revealed a troponin-T concentration of 0.006 µg/L, which was within normal limits (Figure 3). The patient demonstrated satisfactory functional capacity to be able to work at a construction site just before the trauma. Therefore, an additional preoperative cardiac evaluation was not considered necessary.
Figure 2. Chest computed tomography (CT) images showing coronary stents in the main bifurcation of the left coronary arteries (white arrow in (a) and (b)).
Figure 3. Preoperative electrocardiogram of the patient showing normal sinus rhythm with a heart rate of 95 bpm.
bpm, beats per minute.
Immediately after admission to the ICU, the patient’s bladder temperature was 38.0°C. Using a hydrogel-coated water-circulating pad (Arctic Sun®; Medivance Inc., Louisville, CO, USA), which is a surface cooling device, the patient underwent mild therapeutic hypothermia with a target temperature of 33.0°C. During therapeutic hypothermia, sedation and shivering control were performed using continuous rate infusions of midazolam at 5 mg/hour and remifentanil at 0.2 µg/kg/minute. The patient’s bladder temperature was 33.4°C just before he was transferred to the operating room. Preoperative laboratory tests showed no unusual findings except for mildly elevated aspartate aminotransferase (73 U/L) and alanine aminotransferase (51 U/L) concentrations.
The patient’s vital signs upon arrival at the operating room were a BP of 130/70 mmHg and SpO2 of 100%. Lead II of the ECG was monitored, and there were no unusual findings except sinus tachycardia with an HR of 100 bpm on the initial ECG. Arterial BP (ABP), central venous pressure (CVP), and esophageal temperature were monitored from the right radial artery, right subclavian vein, and lower third of the esophagus using a temperature sensor attached to the esophageal stethoscope, respectively. The initial ABP, CVP, and esophageal temperature were 140/85 mmHg, 6 mmHg, and 33.0°C, respectively. Arterial blood gas analysis was performed in the arterial line, with values calculated as for a body temperature of 37°C, which revealed the following: pH, 7.400; partial pressure of carbon dioxide (PCO2), 40 mmHg; partial pressure of oxygen (PO2), 174 mmHg; bicarbonate (HCO3), 24.8 mmol/L; base excess, 0.0; lactate, 0.17 mmol/L; ionized calcium, 1.04 mmol/L; and potassium, 3.3 mmol/L. Total intravenous (IV) anesthesia was performed using a target-controlled infusion of propofol and remifentanil in accordance with the Schnider6 and Minto7 models, respectively. Propofol and remifentanil were infused continuously with effect-site concentrations (Ce) of 1.5 µg/mL and 1.5 ng/mL, respectively. Concurrently, 50 mg of rocuronium was administered by IV bolus injection for neuromuscular blockade. Mechanical ventilation was adjusted to maintain the end-tidal carbon dioxide value at 30 to 33 mmHg.
As ABP decreased to 78/42 mmHg and CVP decreased to 4 mmHg within 10 minutes after anesthesia, the Ce of propofol and remifentanil were lowered to 0.8 µg/mL and 1.0 ng/mL, respectively, and approximately 1 L of balanced crystalloid fluid loading was performed. However, the patient’s BP did not recover, and a dopamine continuous infusion (5 µg/kg/minute) was initiated. When a surgical incision was made 10 minutes after beginning the dopamine infusion, the patient’s ABP and HR increased to 115/70 mmHg and 120 bpm, respectively, and marked ST-segment depression (−2.4 mm) was found, which was not observed on the preoperative or initial ECGs (Figure 4). Subsequently, the depth of anesthesia was increased by increasing the Ce of propofol to 1.0 to 1.2 µg/mL, followed by a continuous infusion of isosorbide dinitrate (1 µg/kg/minute). While HR was maintained at 110 to 120 bpm, ABP gradually decreased to 70/40 mmHg after approximately 5 minutes. To treat the hypotension, isosorbide dinitrate was discontinued, and a bolus injection of 100 µg phenylephrine was administered, after which, ABP increased temporarily to 95/60 mmHg. Subsequently, HR decreased suddenly and rapidly, and ventricular fibrillation occurred. The anesthesiologists immediately discontinued the surgery and performed cardiopulmonary resuscitation (CPR). The patient’s esophageal temperature at the time was 32.0°C.
Figure 4. Intraoperative electrocardiogram of the patient showing sinus tachycardia with significant ST depression. The ST depression on the monitor represented approximately −2.4 at the time.
Chest compressions were performed with periodic defibrillation with 200 J, and epinephrine injections were administered every 3 minutes; these measures elicited no response. During CPR, ECG monitoring showed temporary severe bradycardia (HR, 10–30 bpm). However, this was considered pulseless electrical activity because no pulse was detected in the patient’s femoral artery. Soon afterward, there was a rapid transition to ventricular fibrillation, and CPR was repeated. At the time, the patient’s esophageal temperature dropped to 30.0°C. According to the refractory state after the defibrillation and drug injections, we considered that the cardiac arrest was caused by hypothermia. To reverse the cause, a forced-air warming device and a warming blanket (Bair Hugger®; Arizant Healthcare Inc., Eden Prairie, MN, USA) were used for surface warming, the IV fluids were warmed, and surgical wound closure and dressing were performed rapidly to prevent surface heat loss. The first arterial blood gas analysis after ventricular fibrillation, calculated as for a body temperature of 37°C, showed the following: pH, 7.690; PCO2, 15 mmHg; PO2, 260 mmHg; HCO3, 18.1 mmol/L; base excess, −1.2; lactate, 0.33 mmol/L; ionized calcium, 0.88 mmol/L; and potassium, 5.0 mmol/L. Follow-up arterial blood gas analyses were repeated periodically. As we continued to perform CPR, the patient developed severe metabolic acidosis and hypocalcemia characterized by a pH of 6.840, HCO3 of 15.4 mmol/L, and ionized calcium of 0.66 mmol/L. To correct the abnormal findings, we administered appropriate amounts of sodium bicarbonate and calcium chloride. After an hour-long CPR and rewarming, the patient’s esophageal temperature recovered to 33.0°C, which was followed by observing sinus rhythm characterized by an HR of 75 bpm on an ECG monitor and detecting a pulse in the femoral artery with a characteristic ABP waveform that eventually led to the recovery of BP to 130/55 mmHg. Upon confirming recovery of spontaneous circulation (ROSC), the patient was transported to the ICU while we administered dopamine (10 µg/kg/minute), norepinephrine (0.1 µg/kg/minute), and vasopressin (2 U/hour). ABP, HR, and esophageal temperature during anesthesia are shown in Figure 5.
Figure 5. Changes in vital signs (blood pressure (BP), heart rate (HR), and esophageal temperature) over time (5-minute intervals).
bpm, beats per minute.
Cardiac biomarker testing performed after the patient’s transfer to the ICU showed an increased troponin-T level of 1.432 µg/L. Portable Doppler transthoracic echocardiography (TTE) revealed total occlusion of the left main coronary artery, which had undergone PCI previously. Because emergency coronary artery bypass grafting (CABG) was impossible at the time, repeated PCI was considered an alternative treatment option. However, even after his transfer to the ICU, the patient experienced repetitive ventricular fibrillation. Four hours after the patient’s transfer to the ICU, the patient’s guardian requested that CPR should be stopped. Without resuscitation, the patient died.
Discussion
In this case, the patient, who was receiving therapeutic hypothermia, underwent emergency craniectomy owing to worsening brain hemorrhage, and cardiac arrest occurred during general anesthesia. We deduced that the cardiac arrest was owing to the additional decrease in the patient’s core body temperature and myocardial infarction caused by restenosis of the previous PCI-treated lesion.
Therapeutic hypothermia is used to improve neurological outcomes in cardiac arrest, TBI, stroke, and hypoxic–ischemic encephalopathy.1,4 Studies on the efficacy of therapeutic hypothermia for TBI have been less convincing, and they show conflicting results. However, hypothermia is consistently used as an option for TBI because it reduces intracranial pressure and the cerebral metabolic rate, thereby protecting neuronal function in the brain and decreasing the wide range of injuries or reparative processes by attenuating temperature-sensitive mechanisms, such as excitotoxicity, free radical generation, apoptosis, and inflammation.1,2
The cardiovascular effects associated with hypothermia are complex. Hypothermia affects the electrical conduction of myocardial cells, which can induce hypotension or bradycardia and even increase the risk of ventricular fibrillation by prolonging repolarization.4,8 However, mild hypothermia of 32.0°C to 35.0°C demonstrates cardioprotective effects by decreasing myocardial oxygen demand and increasing myocardial perfusion through vasodilation of coronary vessels.9 Therefore, there has been a growing interest in the clinical use of mild therapeutic hypothermia as an adjunctive therapy during ischemic heart disease.10 Our patient received mild therapeutic hypothermia with a target core temperature of 33.0°C, which may have yielded cardioprotective as well as neuroprotective benefits.11
During general anesthesia, hypothermia is induced following an increase in heat loss caused by peripheral vasodilation after the injection of anesthetic drugs, heat redistribution from the core to the periphery, and impaired normal thermoregulatory responses, such as vasoconstriction or shivering.5 In our patient, the cold operating room temperature and irrigating an uncovered surgical area in the brain with cold saline also may have contributed to the additional core body temperature drop in addition to the effects of general anesthesia.12 However, in our case, neurosurgeons requested that anesthesiologists not perform warming during anesthesia. Instead, esophageal temperature and signs that could result from hypothermia, such as ECG changes, were carefully monitored. The esophageal temperature decreased from 33.0°C, which was measured immediately after entering the operation room, to 32.0°C at the time of cardiac arrest. The esophageal temperature further decreased to a minimum of 30.0°C during CPR. With rewarming, the temperature recovered to 33.0°C and finally, ROSC was achieved. Considering that chest compressions, drugs, and defibrillation did not induce ROSC, which was achieved with the recovery of an esophageal temperature of 33.0°C, the additional rapid fall in the patient’s core body temperature during anesthesia may have caused the cardiac arrest.
In this case study, we suspected that restenosis at the site of the previous stent insertion contributed to the cardiac arrest, given the following three findings: First, the patient had a normal preoperative ECG with no signs of ischemia but developed ST-segment depression during anesthesia. Second, repetitive cardiac arrests occurred even after achieving ROSC with rewarming. Third, an increased troponin-T concentration was observed, and suspected re-stenotic occlusion of the left main coronary artery was confirmed by Doppler TTE after the patient’s transfer to the ICU. Accordingly, myocardial infarction was considered the main cause of death.
The mechanisms of myocardial infarction owing to stent restenosis after DES insertion are rapid stent thrombosis or profuse in-stent restenosis (ISR).13 We assume that the lesion characteristics in the left main bifurcation of the coronary artery and DAPT interruption owing to the increasing amount of cerebral hemorrhage contributed to the stent restenosis in this patient. As evidence in support of this assumption, previous studies have shown that stenting at the left main bifurcation and DAPT interruption are strong risk factors for stent thrombosis or ISR.14,15 While controversial, not interrupting aspirin perioperatively could have avoided poor outcomes.16 However, there were no clinical signs related to expected ongoing restenosis of the coronary artery or myocardial ischemia preoperatively in our patient. We suspect that the possible ischemic signs may have been masked by the aforementioned cardioprotective benefits of therapeutic hypothermia. We also believe that the following three conditions contributed to the intraoperative myocardial ischemia and infarction caused by acute restenosis of the coronary stent: First, intraoperative hypotension may have contributed to myocardial injury. The IV anesthetics used during anesthesia, propofol and remifentanil, induced redistribution of blood volume by peripheral vasodilation, resulting in profound hypotension.4 Profound hypotension was strongly associated with myocardial injury and especially harmful for this patient, who had preexisting ischemic heart disease.17 Second, tachycardia, which may have been caused by the inotrope dopamine, used to compensate for the low BP or the stress response to surgical stimulation from shallow anesthetic depth, may have triggered myocardial ischemia by decreasing oxygen delivery and increasing oxygen consumption.17,18 If cardiogenic shock is suspected, as in our case, norepinephrine could be a better option than dopamine.19 Third, the bolus injection of phenylephrine, a selective α1-adrenergic agonist, just before cardiac arrest may have temporarily and abruptly increased cardiac afterload and impaired myocardial perfusion.20
Had we known that coronary artery occlusion would result from hypothermia-induced cardiac arrest and coronary stent restenosis, the treatment of choice would have been CABG surgery while maintaining circulation with emergency cardiopulmonary bypass (CPB) or extracorporeal membrane oxygenation (ECMO) via the femoral artery and vein.11 However, it was difficult to determine the etiology of the cardiac arrest at the time. Even if we had known the cause, ideal treatment would have been impossible, given circumstances beyond our control; all thoracic surgeons were busy operating, and emergency CPB and ECMO were not available. As an alternative, repeat PCI after the patient achieved ROSC could have been performed. However, the patient had a prolonged CPR duration, and he did not maintain the ROSC state for a period sufficient for intervention, which led to a poor outcome.
In this case, it was not possible to accurately determine the reason for the cardiac arrest and for not achieving consistent ROSC. Instead, we conclude that cardiac arrest developed and persisted as a result of the unintentional severe hypothermia that occurred during general anesthesia and/or coronary artery restenosis at the site of the previous PCI. A thorough evaluation of heart-related history is required for TBI patients who have a history of ischemic heart diseases and who undergo therapeutic hypothermia because there may be undiscovered coronary insufficiency owing to the cardioprotective effects of therapeutic hypothermia. In addition, careful monitoring of the patient’s core body temperature, and preventive and countermeasures should be fully prepared and taken against adverse hemodynamic effects, even cardiac arrest, which can occur because of additive hypothermic responses during anesthesia.
Ethics statement: This case report was approved by our Institutional Review Board (EMC 2020-05-005-001), and the case report was performed in compliance with the EQUATOR Network guidelines (the CARE guidelines). All patient details were de-identified, and written consent was obtained from the patient’s guardian for treatment and publication of this report.
Declaration of conflicting interest: The authors declare that there is no conflict of interest.
Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
ORCID iDs: Dong Ho Park https://orcid.org/0000-0002-6587-3756
Tae Woo Kim https://orcid.org/0000-0002-2115-6448
Mo Se Kim https://orcid.org/0000-0001-7362-0030 | 100 µg (micrograms). | DrugDosage | CC BY-NC | 33499678 | 19,395,621 | 2021-01 |
What was the dosage of drug 'REMIFENTANIL HYDROCHLORIDE'? | Cardiac arrest caused by accidental severe hypothermia and myocardial infarction during general anesthesia.
Therapeutic hypothermia is often used for traumatic brain injury because of its neuroprotective effect and decreased secondary brain injury. However, this procedure lacks clinical evidence supporting its efficacy, and adverse outcomes have been reported during general anesthesia. A 61-year-old man with a history of percutaneous coronary intervention (PCI) was admitted with traumatic brain injury. Immediately after admission, he underwent mild therapeutic hypothermia with a target temperature of 33.0°C for neuroprotection. During general anesthesia for emergency surgery because he developed a mass effect, hypothermic cardiac arrest occurred following an additional decrease in the core body temperature. Moreover, myocardial infarction caused by restenosis of the previous PCI lesion also contributed to the cardiac arrest. Although the patient recovered spontaneous circulation after an hour-long cardiopulmonary resuscitation with rewarming, he eventually died of subsequent repetitive cardiac arrests. When anesthetizing patients undergoing therapeutic hypothermia, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.
Introduction
Therapeutic hypothermia is an attractive strategy for traumatic brain injury (TBI), which demonstrates neuroprotective effects and decreases secondary brain injury.1 Despite the lack of clinical evidence regarding efficacy, therapeutic hypothermia after TBI has been used for at least half a century because of its potential benefits, which are supported by numerous preclinical studies.2,3 However, there have been no reports on the danger of unintentional severe hypothermia during general anesthesia caused by the additional drop in temperature of patients undergoing therapeutic hypothermia. When anesthetizing patients with comorbidities, such as ischemic heart disease, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.4,5
Case report
A 61-year-old man (height, 168 cm; weight, 70 kg) presented to our emergency room (ER) with head trauma after falling on a cement floor from a 5-meter-tall construction site. Upon arrival at the ER, the patient had a Glasgow coma scale score of 11 and the following vital signs: blood pressure (BP), 147/83 mmHg; heart rate (HR), 86 beats per minute (bpm); respiratory rate, 24 breaths per minute; body temperature from the auditory canal, 36.8°C; and percutaneous oxygen saturation (SpO2), 97%. Upon discovering signs of acute subdural hemorrhage, subarachnoid hemorrhage, and intraventricular hemorrhage from a brain computed tomography (CT) scan performed in the ER, we admitted the patient to the intensive care unit (ICU) for close monitoring. After approximately 6 hours, we performed follow-up brain CT because the patient’s mental status had deteriorated from drowsy to stupor. Upon observing a mass effect characterized by an increase in the hemorrhage volume and a left midline shift on the CT images, the patient was scheduled for immediate intubation and emergency decompressive craniectomy (Figure 1).
Figure 1. Brain computed tomography (CT) images taken 6 hours after intensive care unit admission. (a) Axial view showing acute subarachnoid hemorrhage in the peri-mesencephalic cistern (arrow A) and subdural hemorrhage in the right inferior temporal (arrow B) and right parietal (arrow C) regions. (b) Axial view showing acute subdural hemorrhage in the right temporal region (black arrow) and left midline shift.
Although the patient had a history of ischemic heart disease with percutaneous coronary intervention (PCI), it was difficult to obtain accurate information because there was no relevant data in the patient’s medical records from the authors’ hospital, and his drowsy state interfered with detailed interviews. According to the chest CT scan performed immediately after admission, coronary stents had been inserted at the bifurcation site of the left main coronary artery (Figure 2). It also appeared that the inserted stents were drug-eluting stents (DES) considering that he had received dual antiplatelet therapy (DAPT) with aspirin and clopidogrel (Plavix®; Sanofi, Bridgewater, NJ, USA) a switching platelet P2Y12 inhibitor, for approximately 4 months. However, DAPT was discontinued owing to the worsening cerebral hemorrhage after hospitalization; bridging antiplatelet therapy was also not performed. Before the surgery, the patient showed a normal electrocardiogram (ECG), and cardiac biomarker testing revealed a troponin-T concentration of 0.006 µg/L, which was within normal limits (Figure 3). The patient demonstrated satisfactory functional capacity to be able to work at a construction site just before the trauma. Therefore, an additional preoperative cardiac evaluation was not considered necessary.
Figure 2. Chest computed tomography (CT) images showing coronary stents in the main bifurcation of the left coronary arteries (white arrow in (a) and (b)).
Figure 3. Preoperative electrocardiogram of the patient showing normal sinus rhythm with a heart rate of 95 bpm.
bpm, beats per minute.
Immediately after admission to the ICU, the patient’s bladder temperature was 38.0°C. Using a hydrogel-coated water-circulating pad (Arctic Sun®; Medivance Inc., Louisville, CO, USA), which is a surface cooling device, the patient underwent mild therapeutic hypothermia with a target temperature of 33.0°C. During therapeutic hypothermia, sedation and shivering control were performed using continuous rate infusions of midazolam at 5 mg/hour and remifentanil at 0.2 µg/kg/minute. The patient’s bladder temperature was 33.4°C just before he was transferred to the operating room. Preoperative laboratory tests showed no unusual findings except for mildly elevated aspartate aminotransferase (73 U/L) and alanine aminotransferase (51 U/L) concentrations.
The patient’s vital signs upon arrival at the operating room were a BP of 130/70 mmHg and SpO2 of 100%. Lead II of the ECG was monitored, and there were no unusual findings except sinus tachycardia with an HR of 100 bpm on the initial ECG. Arterial BP (ABP), central venous pressure (CVP), and esophageal temperature were monitored from the right radial artery, right subclavian vein, and lower third of the esophagus using a temperature sensor attached to the esophageal stethoscope, respectively. The initial ABP, CVP, and esophageal temperature were 140/85 mmHg, 6 mmHg, and 33.0°C, respectively. Arterial blood gas analysis was performed in the arterial line, with values calculated as for a body temperature of 37°C, which revealed the following: pH, 7.400; partial pressure of carbon dioxide (PCO2), 40 mmHg; partial pressure of oxygen (PO2), 174 mmHg; bicarbonate (HCO3), 24.8 mmol/L; base excess, 0.0; lactate, 0.17 mmol/L; ionized calcium, 1.04 mmol/L; and potassium, 3.3 mmol/L. Total intravenous (IV) anesthesia was performed using a target-controlled infusion of propofol and remifentanil in accordance with the Schnider6 and Minto7 models, respectively. Propofol and remifentanil were infused continuously with effect-site concentrations (Ce) of 1.5 µg/mL and 1.5 ng/mL, respectively. Concurrently, 50 mg of rocuronium was administered by IV bolus injection for neuromuscular blockade. Mechanical ventilation was adjusted to maintain the end-tidal carbon dioxide value at 30 to 33 mmHg.
As ABP decreased to 78/42 mmHg and CVP decreased to 4 mmHg within 10 minutes after anesthesia, the Ce of propofol and remifentanil were lowered to 0.8 µg/mL and 1.0 ng/mL, respectively, and approximately 1 L of balanced crystalloid fluid loading was performed. However, the patient’s BP did not recover, and a dopamine continuous infusion (5 µg/kg/minute) was initiated. When a surgical incision was made 10 minutes after beginning the dopamine infusion, the patient’s ABP and HR increased to 115/70 mmHg and 120 bpm, respectively, and marked ST-segment depression (−2.4 mm) was found, which was not observed on the preoperative or initial ECGs (Figure 4). Subsequently, the depth of anesthesia was increased by increasing the Ce of propofol to 1.0 to 1.2 µg/mL, followed by a continuous infusion of isosorbide dinitrate (1 µg/kg/minute). While HR was maintained at 110 to 120 bpm, ABP gradually decreased to 70/40 mmHg after approximately 5 minutes. To treat the hypotension, isosorbide dinitrate was discontinued, and a bolus injection of 100 µg phenylephrine was administered, after which, ABP increased temporarily to 95/60 mmHg. Subsequently, HR decreased suddenly and rapidly, and ventricular fibrillation occurred. The anesthesiologists immediately discontinued the surgery and performed cardiopulmonary resuscitation (CPR). The patient’s esophageal temperature at the time was 32.0°C.
Figure 4. Intraoperative electrocardiogram of the patient showing sinus tachycardia with significant ST depression. The ST depression on the monitor represented approximately −2.4 at the time.
Chest compressions were performed with periodic defibrillation with 200 J, and epinephrine injections were administered every 3 minutes; these measures elicited no response. During CPR, ECG monitoring showed temporary severe bradycardia (HR, 10–30 bpm). However, this was considered pulseless electrical activity because no pulse was detected in the patient’s femoral artery. Soon afterward, there was a rapid transition to ventricular fibrillation, and CPR was repeated. At the time, the patient’s esophageal temperature dropped to 30.0°C. According to the refractory state after the defibrillation and drug injections, we considered that the cardiac arrest was caused by hypothermia. To reverse the cause, a forced-air warming device and a warming blanket (Bair Hugger®; Arizant Healthcare Inc., Eden Prairie, MN, USA) were used for surface warming, the IV fluids were warmed, and surgical wound closure and dressing were performed rapidly to prevent surface heat loss. The first arterial blood gas analysis after ventricular fibrillation, calculated as for a body temperature of 37°C, showed the following: pH, 7.690; PCO2, 15 mmHg; PO2, 260 mmHg; HCO3, 18.1 mmol/L; base excess, −1.2; lactate, 0.33 mmol/L; ionized calcium, 0.88 mmol/L; and potassium, 5.0 mmol/L. Follow-up arterial blood gas analyses were repeated periodically. As we continued to perform CPR, the patient developed severe metabolic acidosis and hypocalcemia characterized by a pH of 6.840, HCO3 of 15.4 mmol/L, and ionized calcium of 0.66 mmol/L. To correct the abnormal findings, we administered appropriate amounts of sodium bicarbonate and calcium chloride. After an hour-long CPR and rewarming, the patient’s esophageal temperature recovered to 33.0°C, which was followed by observing sinus rhythm characterized by an HR of 75 bpm on an ECG monitor and detecting a pulse in the femoral artery with a characteristic ABP waveform that eventually led to the recovery of BP to 130/55 mmHg. Upon confirming recovery of spontaneous circulation (ROSC), the patient was transported to the ICU while we administered dopamine (10 µg/kg/minute), norepinephrine (0.1 µg/kg/minute), and vasopressin (2 U/hour). ABP, HR, and esophageal temperature during anesthesia are shown in Figure 5.
Figure 5. Changes in vital signs (blood pressure (BP), heart rate (HR), and esophageal temperature) over time (5-minute intervals).
bpm, beats per minute.
Cardiac biomarker testing performed after the patient’s transfer to the ICU showed an increased troponin-T level of 1.432 µg/L. Portable Doppler transthoracic echocardiography (TTE) revealed total occlusion of the left main coronary artery, which had undergone PCI previously. Because emergency coronary artery bypass grafting (CABG) was impossible at the time, repeated PCI was considered an alternative treatment option. However, even after his transfer to the ICU, the patient experienced repetitive ventricular fibrillation. Four hours after the patient’s transfer to the ICU, the patient’s guardian requested that CPR should be stopped. Without resuscitation, the patient died.
Discussion
In this case, the patient, who was receiving therapeutic hypothermia, underwent emergency craniectomy owing to worsening brain hemorrhage, and cardiac arrest occurred during general anesthesia. We deduced that the cardiac arrest was owing to the additional decrease in the patient’s core body temperature and myocardial infarction caused by restenosis of the previous PCI-treated lesion.
Therapeutic hypothermia is used to improve neurological outcomes in cardiac arrest, TBI, stroke, and hypoxic–ischemic encephalopathy.1,4 Studies on the efficacy of therapeutic hypothermia for TBI have been less convincing, and they show conflicting results. However, hypothermia is consistently used as an option for TBI because it reduces intracranial pressure and the cerebral metabolic rate, thereby protecting neuronal function in the brain and decreasing the wide range of injuries or reparative processes by attenuating temperature-sensitive mechanisms, such as excitotoxicity, free radical generation, apoptosis, and inflammation.1,2
The cardiovascular effects associated with hypothermia are complex. Hypothermia affects the electrical conduction of myocardial cells, which can induce hypotension or bradycardia and even increase the risk of ventricular fibrillation by prolonging repolarization.4,8 However, mild hypothermia of 32.0°C to 35.0°C demonstrates cardioprotective effects by decreasing myocardial oxygen demand and increasing myocardial perfusion through vasodilation of coronary vessels.9 Therefore, there has been a growing interest in the clinical use of mild therapeutic hypothermia as an adjunctive therapy during ischemic heart disease.10 Our patient received mild therapeutic hypothermia with a target core temperature of 33.0°C, which may have yielded cardioprotective as well as neuroprotective benefits.11
During general anesthesia, hypothermia is induced following an increase in heat loss caused by peripheral vasodilation after the injection of anesthetic drugs, heat redistribution from the core to the periphery, and impaired normal thermoregulatory responses, such as vasoconstriction or shivering.5 In our patient, the cold operating room temperature and irrigating an uncovered surgical area in the brain with cold saline also may have contributed to the additional core body temperature drop in addition to the effects of general anesthesia.12 However, in our case, neurosurgeons requested that anesthesiologists not perform warming during anesthesia. Instead, esophageal temperature and signs that could result from hypothermia, such as ECG changes, were carefully monitored. The esophageal temperature decreased from 33.0°C, which was measured immediately after entering the operation room, to 32.0°C at the time of cardiac arrest. The esophageal temperature further decreased to a minimum of 30.0°C during CPR. With rewarming, the temperature recovered to 33.0°C and finally, ROSC was achieved. Considering that chest compressions, drugs, and defibrillation did not induce ROSC, which was achieved with the recovery of an esophageal temperature of 33.0°C, the additional rapid fall in the patient’s core body temperature during anesthesia may have caused the cardiac arrest.
In this case study, we suspected that restenosis at the site of the previous stent insertion contributed to the cardiac arrest, given the following three findings: First, the patient had a normal preoperative ECG with no signs of ischemia but developed ST-segment depression during anesthesia. Second, repetitive cardiac arrests occurred even after achieving ROSC with rewarming. Third, an increased troponin-T concentration was observed, and suspected re-stenotic occlusion of the left main coronary artery was confirmed by Doppler TTE after the patient’s transfer to the ICU. Accordingly, myocardial infarction was considered the main cause of death.
The mechanisms of myocardial infarction owing to stent restenosis after DES insertion are rapid stent thrombosis or profuse in-stent restenosis (ISR).13 We assume that the lesion characteristics in the left main bifurcation of the coronary artery and DAPT interruption owing to the increasing amount of cerebral hemorrhage contributed to the stent restenosis in this patient. As evidence in support of this assumption, previous studies have shown that stenting at the left main bifurcation and DAPT interruption are strong risk factors for stent thrombosis or ISR.14,15 While controversial, not interrupting aspirin perioperatively could have avoided poor outcomes.16 However, there were no clinical signs related to expected ongoing restenosis of the coronary artery or myocardial ischemia preoperatively in our patient. We suspect that the possible ischemic signs may have been masked by the aforementioned cardioprotective benefits of therapeutic hypothermia. We also believe that the following three conditions contributed to the intraoperative myocardial ischemia and infarction caused by acute restenosis of the coronary stent: First, intraoperative hypotension may have contributed to myocardial injury. The IV anesthetics used during anesthesia, propofol and remifentanil, induced redistribution of blood volume by peripheral vasodilation, resulting in profound hypotension.4 Profound hypotension was strongly associated with myocardial injury and especially harmful for this patient, who had preexisting ischemic heart disease.17 Second, tachycardia, which may have been caused by the inotrope dopamine, used to compensate for the low BP or the stress response to surgical stimulation from shallow anesthetic depth, may have triggered myocardial ischemia by decreasing oxygen delivery and increasing oxygen consumption.17,18 If cardiogenic shock is suspected, as in our case, norepinephrine could be a better option than dopamine.19 Third, the bolus injection of phenylephrine, a selective α1-adrenergic agonist, just before cardiac arrest may have temporarily and abruptly increased cardiac afterload and impaired myocardial perfusion.20
Had we known that coronary artery occlusion would result from hypothermia-induced cardiac arrest and coronary stent restenosis, the treatment of choice would have been CABG surgery while maintaining circulation with emergency cardiopulmonary bypass (CPB) or extracorporeal membrane oxygenation (ECMO) via the femoral artery and vein.11 However, it was difficult to determine the etiology of the cardiac arrest at the time. Even if we had known the cause, ideal treatment would have been impossible, given circumstances beyond our control; all thoracic surgeons were busy operating, and emergency CPB and ECMO were not available. As an alternative, repeat PCI after the patient achieved ROSC could have been performed. However, the patient had a prolonged CPR duration, and he did not maintain the ROSC state for a period sufficient for intervention, which led to a poor outcome.
In this case, it was not possible to accurately determine the reason for the cardiac arrest and for not achieving consistent ROSC. Instead, we conclude that cardiac arrest developed and persisted as a result of the unintentional severe hypothermia that occurred during general anesthesia and/or coronary artery restenosis at the site of the previous PCI. A thorough evaluation of heart-related history is required for TBI patients who have a history of ischemic heart diseases and who undergo therapeutic hypothermia because there may be undiscovered coronary insufficiency owing to the cardioprotective effects of therapeutic hypothermia. In addition, careful monitoring of the patient’s core body temperature, and preventive and countermeasures should be fully prepared and taken against adverse hemodynamic effects, even cardiac arrest, which can occur because of additive hypothermic responses during anesthesia.
Ethics statement: This case report was approved by our Institutional Review Board (EMC 2020-05-005-001), and the case report was performed in compliance with the EQUATOR Network guidelines (the CARE guidelines). All patient details were de-identified, and written consent was obtained from the patient’s guardian for treatment and publication of this report.
Declaration of conflicting interest: The authors declare that there is no conflict of interest.
Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
ORCID iDs: Dong Ho Park https://orcid.org/0000-0002-6587-3756
Tae Woo Kim https://orcid.org/0000-0002-2115-6448
Mo Se Kim https://orcid.org/0000-0001-7362-0030 | EFFECT?SITE CONCENTRATIONS (CE) OF 1.5 UG/ML | DrugDosageText | CC BY-NC | 33499678 | 19,429,456 | 2021-01 |
What was the dosage of drug 'REMIFENTANIL'? | Cardiac arrest caused by accidental severe hypothermia and myocardial infarction during general anesthesia.
Therapeutic hypothermia is often used for traumatic brain injury because of its neuroprotective effect and decreased secondary brain injury. However, this procedure lacks clinical evidence supporting its efficacy, and adverse outcomes have been reported during general anesthesia. A 61-year-old man with a history of percutaneous coronary intervention (PCI) was admitted with traumatic brain injury. Immediately after admission, he underwent mild therapeutic hypothermia with a target temperature of 33.0°C for neuroprotection. During general anesthesia for emergency surgery because he developed a mass effect, hypothermic cardiac arrest occurred following an additional decrease in the core body temperature. Moreover, myocardial infarction caused by restenosis of the previous PCI lesion also contributed to the cardiac arrest. Although the patient recovered spontaneous circulation after an hour-long cardiopulmonary resuscitation with rewarming, he eventually died of subsequent repetitive cardiac arrests. When anesthetizing patients undergoing therapeutic hypothermia, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.
Introduction
Therapeutic hypothermia is an attractive strategy for traumatic brain injury (TBI), which demonstrates neuroprotective effects and decreases secondary brain injury.1 Despite the lack of clinical evidence regarding efficacy, therapeutic hypothermia after TBI has been used for at least half a century because of its potential benefits, which are supported by numerous preclinical studies.2,3 However, there have been no reports on the danger of unintentional severe hypothermia during general anesthesia caused by the additional drop in temperature of patients undergoing therapeutic hypothermia. When anesthetizing patients with comorbidities, such as ischemic heart disease, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.4,5
Case report
A 61-year-old man (height, 168 cm; weight, 70 kg) presented to our emergency room (ER) with head trauma after falling on a cement floor from a 5-meter-tall construction site. Upon arrival at the ER, the patient had a Glasgow coma scale score of 11 and the following vital signs: blood pressure (BP), 147/83 mmHg; heart rate (HR), 86 beats per minute (bpm); respiratory rate, 24 breaths per minute; body temperature from the auditory canal, 36.8°C; and percutaneous oxygen saturation (SpO2), 97%. Upon discovering signs of acute subdural hemorrhage, subarachnoid hemorrhage, and intraventricular hemorrhage from a brain computed tomography (CT) scan performed in the ER, we admitted the patient to the intensive care unit (ICU) for close monitoring. After approximately 6 hours, we performed follow-up brain CT because the patient’s mental status had deteriorated from drowsy to stupor. Upon observing a mass effect characterized by an increase in the hemorrhage volume and a left midline shift on the CT images, the patient was scheduled for immediate intubation and emergency decompressive craniectomy (Figure 1).
Figure 1. Brain computed tomography (CT) images taken 6 hours after intensive care unit admission. (a) Axial view showing acute subarachnoid hemorrhage in the peri-mesencephalic cistern (arrow A) and subdural hemorrhage in the right inferior temporal (arrow B) and right parietal (arrow C) regions. (b) Axial view showing acute subdural hemorrhage in the right temporal region (black arrow) and left midline shift.
Although the patient had a history of ischemic heart disease with percutaneous coronary intervention (PCI), it was difficult to obtain accurate information because there was no relevant data in the patient’s medical records from the authors’ hospital, and his drowsy state interfered with detailed interviews. According to the chest CT scan performed immediately after admission, coronary stents had been inserted at the bifurcation site of the left main coronary artery (Figure 2). It also appeared that the inserted stents were drug-eluting stents (DES) considering that he had received dual antiplatelet therapy (DAPT) with aspirin and clopidogrel (Plavix®; Sanofi, Bridgewater, NJ, USA) a switching platelet P2Y12 inhibitor, for approximately 4 months. However, DAPT was discontinued owing to the worsening cerebral hemorrhage after hospitalization; bridging antiplatelet therapy was also not performed. Before the surgery, the patient showed a normal electrocardiogram (ECG), and cardiac biomarker testing revealed a troponin-T concentration of 0.006 µg/L, which was within normal limits (Figure 3). The patient demonstrated satisfactory functional capacity to be able to work at a construction site just before the trauma. Therefore, an additional preoperative cardiac evaluation was not considered necessary.
Figure 2. Chest computed tomography (CT) images showing coronary stents in the main bifurcation of the left coronary arteries (white arrow in (a) and (b)).
Figure 3. Preoperative electrocardiogram of the patient showing normal sinus rhythm with a heart rate of 95 bpm.
bpm, beats per minute.
Immediately after admission to the ICU, the patient’s bladder temperature was 38.0°C. Using a hydrogel-coated water-circulating pad (Arctic Sun®; Medivance Inc., Louisville, CO, USA), which is a surface cooling device, the patient underwent mild therapeutic hypothermia with a target temperature of 33.0°C. During therapeutic hypothermia, sedation and shivering control were performed using continuous rate infusions of midazolam at 5 mg/hour and remifentanil at 0.2 µg/kg/minute. The patient’s bladder temperature was 33.4°C just before he was transferred to the operating room. Preoperative laboratory tests showed no unusual findings except for mildly elevated aspartate aminotransferase (73 U/L) and alanine aminotransferase (51 U/L) concentrations.
The patient’s vital signs upon arrival at the operating room were a BP of 130/70 mmHg and SpO2 of 100%. Lead II of the ECG was monitored, and there were no unusual findings except sinus tachycardia with an HR of 100 bpm on the initial ECG. Arterial BP (ABP), central venous pressure (CVP), and esophageal temperature were monitored from the right radial artery, right subclavian vein, and lower third of the esophagus using a temperature sensor attached to the esophageal stethoscope, respectively. The initial ABP, CVP, and esophageal temperature were 140/85 mmHg, 6 mmHg, and 33.0°C, respectively. Arterial blood gas analysis was performed in the arterial line, with values calculated as for a body temperature of 37°C, which revealed the following: pH, 7.400; partial pressure of carbon dioxide (PCO2), 40 mmHg; partial pressure of oxygen (PO2), 174 mmHg; bicarbonate (HCO3), 24.8 mmol/L; base excess, 0.0; lactate, 0.17 mmol/L; ionized calcium, 1.04 mmol/L; and potassium, 3.3 mmol/L. Total intravenous (IV) anesthesia was performed using a target-controlled infusion of propofol and remifentanil in accordance with the Schnider6 and Minto7 models, respectively. Propofol and remifentanil were infused continuously with effect-site concentrations (Ce) of 1.5 µg/mL and 1.5 ng/mL, respectively. Concurrently, 50 mg of rocuronium was administered by IV bolus injection for neuromuscular blockade. Mechanical ventilation was adjusted to maintain the end-tidal carbon dioxide value at 30 to 33 mmHg.
As ABP decreased to 78/42 mmHg and CVP decreased to 4 mmHg within 10 minutes after anesthesia, the Ce of propofol and remifentanil were lowered to 0.8 µg/mL and 1.0 ng/mL, respectively, and approximately 1 L of balanced crystalloid fluid loading was performed. However, the patient’s BP did not recover, and a dopamine continuous infusion (5 µg/kg/minute) was initiated. When a surgical incision was made 10 minutes after beginning the dopamine infusion, the patient’s ABP and HR increased to 115/70 mmHg and 120 bpm, respectively, and marked ST-segment depression (−2.4 mm) was found, which was not observed on the preoperative or initial ECGs (Figure 4). Subsequently, the depth of anesthesia was increased by increasing the Ce of propofol to 1.0 to 1.2 µg/mL, followed by a continuous infusion of isosorbide dinitrate (1 µg/kg/minute). While HR was maintained at 110 to 120 bpm, ABP gradually decreased to 70/40 mmHg after approximately 5 minutes. To treat the hypotension, isosorbide dinitrate was discontinued, and a bolus injection of 100 µg phenylephrine was administered, after which, ABP increased temporarily to 95/60 mmHg. Subsequently, HR decreased suddenly and rapidly, and ventricular fibrillation occurred. The anesthesiologists immediately discontinued the surgery and performed cardiopulmonary resuscitation (CPR). The patient’s esophageal temperature at the time was 32.0°C.
Figure 4. Intraoperative electrocardiogram of the patient showing sinus tachycardia with significant ST depression. The ST depression on the monitor represented approximately −2.4 at the time.
Chest compressions were performed with periodic defibrillation with 200 J, and epinephrine injections were administered every 3 minutes; these measures elicited no response. During CPR, ECG monitoring showed temporary severe bradycardia (HR, 10–30 bpm). However, this was considered pulseless electrical activity because no pulse was detected in the patient’s femoral artery. Soon afterward, there was a rapid transition to ventricular fibrillation, and CPR was repeated. At the time, the patient’s esophageal temperature dropped to 30.0°C. According to the refractory state after the defibrillation and drug injections, we considered that the cardiac arrest was caused by hypothermia. To reverse the cause, a forced-air warming device and a warming blanket (Bair Hugger®; Arizant Healthcare Inc., Eden Prairie, MN, USA) were used for surface warming, the IV fluids were warmed, and surgical wound closure and dressing were performed rapidly to prevent surface heat loss. The first arterial blood gas analysis after ventricular fibrillation, calculated as for a body temperature of 37°C, showed the following: pH, 7.690; PCO2, 15 mmHg; PO2, 260 mmHg; HCO3, 18.1 mmol/L; base excess, −1.2; lactate, 0.33 mmol/L; ionized calcium, 0.88 mmol/L; and potassium, 5.0 mmol/L. Follow-up arterial blood gas analyses were repeated periodically. As we continued to perform CPR, the patient developed severe metabolic acidosis and hypocalcemia characterized by a pH of 6.840, HCO3 of 15.4 mmol/L, and ionized calcium of 0.66 mmol/L. To correct the abnormal findings, we administered appropriate amounts of sodium bicarbonate and calcium chloride. After an hour-long CPR and rewarming, the patient’s esophageal temperature recovered to 33.0°C, which was followed by observing sinus rhythm characterized by an HR of 75 bpm on an ECG monitor and detecting a pulse in the femoral artery with a characteristic ABP waveform that eventually led to the recovery of BP to 130/55 mmHg. Upon confirming recovery of spontaneous circulation (ROSC), the patient was transported to the ICU while we administered dopamine (10 µg/kg/minute), norepinephrine (0.1 µg/kg/minute), and vasopressin (2 U/hour). ABP, HR, and esophageal temperature during anesthesia are shown in Figure 5.
Figure 5. Changes in vital signs (blood pressure (BP), heart rate (HR), and esophageal temperature) over time (5-minute intervals).
bpm, beats per minute.
Cardiac biomarker testing performed after the patient’s transfer to the ICU showed an increased troponin-T level of 1.432 µg/L. Portable Doppler transthoracic echocardiography (TTE) revealed total occlusion of the left main coronary artery, which had undergone PCI previously. Because emergency coronary artery bypass grafting (CABG) was impossible at the time, repeated PCI was considered an alternative treatment option. However, even after his transfer to the ICU, the patient experienced repetitive ventricular fibrillation. Four hours after the patient’s transfer to the ICU, the patient’s guardian requested that CPR should be stopped. Without resuscitation, the patient died.
Discussion
In this case, the patient, who was receiving therapeutic hypothermia, underwent emergency craniectomy owing to worsening brain hemorrhage, and cardiac arrest occurred during general anesthesia. We deduced that the cardiac arrest was owing to the additional decrease in the patient’s core body temperature and myocardial infarction caused by restenosis of the previous PCI-treated lesion.
Therapeutic hypothermia is used to improve neurological outcomes in cardiac arrest, TBI, stroke, and hypoxic–ischemic encephalopathy.1,4 Studies on the efficacy of therapeutic hypothermia for TBI have been less convincing, and they show conflicting results. However, hypothermia is consistently used as an option for TBI because it reduces intracranial pressure and the cerebral metabolic rate, thereby protecting neuronal function in the brain and decreasing the wide range of injuries or reparative processes by attenuating temperature-sensitive mechanisms, such as excitotoxicity, free radical generation, apoptosis, and inflammation.1,2
The cardiovascular effects associated with hypothermia are complex. Hypothermia affects the electrical conduction of myocardial cells, which can induce hypotension or bradycardia and even increase the risk of ventricular fibrillation by prolonging repolarization.4,8 However, mild hypothermia of 32.0°C to 35.0°C demonstrates cardioprotective effects by decreasing myocardial oxygen demand and increasing myocardial perfusion through vasodilation of coronary vessels.9 Therefore, there has been a growing interest in the clinical use of mild therapeutic hypothermia as an adjunctive therapy during ischemic heart disease.10 Our patient received mild therapeutic hypothermia with a target core temperature of 33.0°C, which may have yielded cardioprotective as well as neuroprotective benefits.11
During general anesthesia, hypothermia is induced following an increase in heat loss caused by peripheral vasodilation after the injection of anesthetic drugs, heat redistribution from the core to the periphery, and impaired normal thermoregulatory responses, such as vasoconstriction or shivering.5 In our patient, the cold operating room temperature and irrigating an uncovered surgical area in the brain with cold saline also may have contributed to the additional core body temperature drop in addition to the effects of general anesthesia.12 However, in our case, neurosurgeons requested that anesthesiologists not perform warming during anesthesia. Instead, esophageal temperature and signs that could result from hypothermia, such as ECG changes, were carefully monitored. The esophageal temperature decreased from 33.0°C, which was measured immediately after entering the operation room, to 32.0°C at the time of cardiac arrest. The esophageal temperature further decreased to a minimum of 30.0°C during CPR. With rewarming, the temperature recovered to 33.0°C and finally, ROSC was achieved. Considering that chest compressions, drugs, and defibrillation did not induce ROSC, which was achieved with the recovery of an esophageal temperature of 33.0°C, the additional rapid fall in the patient’s core body temperature during anesthesia may have caused the cardiac arrest.
In this case study, we suspected that restenosis at the site of the previous stent insertion contributed to the cardiac arrest, given the following three findings: First, the patient had a normal preoperative ECG with no signs of ischemia but developed ST-segment depression during anesthesia. Second, repetitive cardiac arrests occurred even after achieving ROSC with rewarming. Third, an increased troponin-T concentration was observed, and suspected re-stenotic occlusion of the left main coronary artery was confirmed by Doppler TTE after the patient’s transfer to the ICU. Accordingly, myocardial infarction was considered the main cause of death.
The mechanisms of myocardial infarction owing to stent restenosis after DES insertion are rapid stent thrombosis or profuse in-stent restenosis (ISR).13 We assume that the lesion characteristics in the left main bifurcation of the coronary artery and DAPT interruption owing to the increasing amount of cerebral hemorrhage contributed to the stent restenosis in this patient. As evidence in support of this assumption, previous studies have shown that stenting at the left main bifurcation and DAPT interruption are strong risk factors for stent thrombosis or ISR.14,15 While controversial, not interrupting aspirin perioperatively could have avoided poor outcomes.16 However, there were no clinical signs related to expected ongoing restenosis of the coronary artery or myocardial ischemia preoperatively in our patient. We suspect that the possible ischemic signs may have been masked by the aforementioned cardioprotective benefits of therapeutic hypothermia. We also believe that the following three conditions contributed to the intraoperative myocardial ischemia and infarction caused by acute restenosis of the coronary stent: First, intraoperative hypotension may have contributed to myocardial injury. The IV anesthetics used during anesthesia, propofol and remifentanil, induced redistribution of blood volume by peripheral vasodilation, resulting in profound hypotension.4 Profound hypotension was strongly associated with myocardial injury and especially harmful for this patient, who had preexisting ischemic heart disease.17 Second, tachycardia, which may have been caused by the inotrope dopamine, used to compensate for the low BP or the stress response to surgical stimulation from shallow anesthetic depth, may have triggered myocardial ischemia by decreasing oxygen delivery and increasing oxygen consumption.17,18 If cardiogenic shock is suspected, as in our case, norepinephrine could be a better option than dopamine.19 Third, the bolus injection of phenylephrine, a selective α1-adrenergic agonist, just before cardiac arrest may have temporarily and abruptly increased cardiac afterload and impaired myocardial perfusion.20
Had we known that coronary artery occlusion would result from hypothermia-induced cardiac arrest and coronary stent restenosis, the treatment of choice would have been CABG surgery while maintaining circulation with emergency cardiopulmonary bypass (CPB) or extracorporeal membrane oxygenation (ECMO) via the femoral artery and vein.11 However, it was difficult to determine the etiology of the cardiac arrest at the time. Even if we had known the cause, ideal treatment would have been impossible, given circumstances beyond our control; all thoracic surgeons were busy operating, and emergency CPB and ECMO were not available. As an alternative, repeat PCI after the patient achieved ROSC could have been performed. However, the patient had a prolonged CPR duration, and he did not maintain the ROSC state for a period sufficient for intervention, which led to a poor outcome.
In this case, it was not possible to accurately determine the reason for the cardiac arrest and for not achieving consistent ROSC. Instead, we conclude that cardiac arrest developed and persisted as a result of the unintentional severe hypothermia that occurred during general anesthesia and/or coronary artery restenosis at the site of the previous PCI. A thorough evaluation of heart-related history is required for TBI patients who have a history of ischemic heart diseases and who undergo therapeutic hypothermia because there may be undiscovered coronary insufficiency owing to the cardioprotective effects of therapeutic hypothermia. In addition, careful monitoring of the patient’s core body temperature, and preventive and countermeasures should be fully prepared and taken against adverse hemodynamic effects, even cardiac arrest, which can occur because of additive hypothermic responses during anesthesia.
Ethics statement: This case report was approved by our Institutional Review Board (EMC 2020-05-005-001), and the case report was performed in compliance with the EQUATOR Network guidelines (the CARE guidelines). All patient details were de-identified, and written consent was obtained from the patient’s guardian for treatment and publication of this report.
Declaration of conflicting interest: The authors declare that there is no conflict of interest.
Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
ORCID iDs: Dong Ho Park https://orcid.org/0000-0002-6587-3756
Tae Woo Kim https://orcid.org/0000-0002-2115-6448
Mo Se Kim https://orcid.org/0000-0001-7362-0030 | EFFECT?SITE CONCENTRATIONS (CE) OF 0.8 UG/ML | DrugDosageText | CC BY-NC | 33499678 | 19,429,456 | 2021-01 |
What was the dosage of drug 'ROCURONIUM BROMIDE'? | Cardiac arrest caused by accidental severe hypothermia and myocardial infarction during general anesthesia.
Therapeutic hypothermia is often used for traumatic brain injury because of its neuroprotective effect and decreased secondary brain injury. However, this procedure lacks clinical evidence supporting its efficacy, and adverse outcomes have been reported during general anesthesia. A 61-year-old man with a history of percutaneous coronary intervention (PCI) was admitted with traumatic brain injury. Immediately after admission, he underwent mild therapeutic hypothermia with a target temperature of 33.0°C for neuroprotection. During general anesthesia for emergency surgery because he developed a mass effect, hypothermic cardiac arrest occurred following an additional decrease in the core body temperature. Moreover, myocardial infarction caused by restenosis of the previous PCI lesion also contributed to the cardiac arrest. Although the patient recovered spontaneous circulation after an hour-long cardiopulmonary resuscitation with rewarming, he eventually died of subsequent repetitive cardiac arrests. When anesthetizing patients undergoing therapeutic hypothermia, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.
Introduction
Therapeutic hypothermia is an attractive strategy for traumatic brain injury (TBI), which demonstrates neuroprotective effects and decreases secondary brain injury.1 Despite the lack of clinical evidence regarding efficacy, therapeutic hypothermia after TBI has been used for at least half a century because of its potential benefits, which are supported by numerous preclinical studies.2,3 However, there have been no reports on the danger of unintentional severe hypothermia during general anesthesia caused by the additional drop in temperature of patients undergoing therapeutic hypothermia. When anesthetizing patients with comorbidities, such as ischemic heart disease, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.4,5
Case report
A 61-year-old man (height, 168 cm; weight, 70 kg) presented to our emergency room (ER) with head trauma after falling on a cement floor from a 5-meter-tall construction site. Upon arrival at the ER, the patient had a Glasgow coma scale score of 11 and the following vital signs: blood pressure (BP), 147/83 mmHg; heart rate (HR), 86 beats per minute (bpm); respiratory rate, 24 breaths per minute; body temperature from the auditory canal, 36.8°C; and percutaneous oxygen saturation (SpO2), 97%. Upon discovering signs of acute subdural hemorrhage, subarachnoid hemorrhage, and intraventricular hemorrhage from a brain computed tomography (CT) scan performed in the ER, we admitted the patient to the intensive care unit (ICU) for close monitoring. After approximately 6 hours, we performed follow-up brain CT because the patient’s mental status had deteriorated from drowsy to stupor. Upon observing a mass effect characterized by an increase in the hemorrhage volume and a left midline shift on the CT images, the patient was scheduled for immediate intubation and emergency decompressive craniectomy (Figure 1).
Figure 1. Brain computed tomography (CT) images taken 6 hours after intensive care unit admission. (a) Axial view showing acute subarachnoid hemorrhage in the peri-mesencephalic cistern (arrow A) and subdural hemorrhage in the right inferior temporal (arrow B) and right parietal (arrow C) regions. (b) Axial view showing acute subdural hemorrhage in the right temporal region (black arrow) and left midline shift.
Although the patient had a history of ischemic heart disease with percutaneous coronary intervention (PCI), it was difficult to obtain accurate information because there was no relevant data in the patient’s medical records from the authors’ hospital, and his drowsy state interfered with detailed interviews. According to the chest CT scan performed immediately after admission, coronary stents had been inserted at the bifurcation site of the left main coronary artery (Figure 2). It also appeared that the inserted stents were drug-eluting stents (DES) considering that he had received dual antiplatelet therapy (DAPT) with aspirin and clopidogrel (Plavix®; Sanofi, Bridgewater, NJ, USA) a switching platelet P2Y12 inhibitor, for approximately 4 months. However, DAPT was discontinued owing to the worsening cerebral hemorrhage after hospitalization; bridging antiplatelet therapy was also not performed. Before the surgery, the patient showed a normal electrocardiogram (ECG), and cardiac biomarker testing revealed a troponin-T concentration of 0.006 µg/L, which was within normal limits (Figure 3). The patient demonstrated satisfactory functional capacity to be able to work at a construction site just before the trauma. Therefore, an additional preoperative cardiac evaluation was not considered necessary.
Figure 2. Chest computed tomography (CT) images showing coronary stents in the main bifurcation of the left coronary arteries (white arrow in (a) and (b)).
Figure 3. Preoperative electrocardiogram of the patient showing normal sinus rhythm with a heart rate of 95 bpm.
bpm, beats per minute.
Immediately after admission to the ICU, the patient’s bladder temperature was 38.0°C. Using a hydrogel-coated water-circulating pad (Arctic Sun®; Medivance Inc., Louisville, CO, USA), which is a surface cooling device, the patient underwent mild therapeutic hypothermia with a target temperature of 33.0°C. During therapeutic hypothermia, sedation and shivering control were performed using continuous rate infusions of midazolam at 5 mg/hour and remifentanil at 0.2 µg/kg/minute. The patient’s bladder temperature was 33.4°C just before he was transferred to the operating room. Preoperative laboratory tests showed no unusual findings except for mildly elevated aspartate aminotransferase (73 U/L) and alanine aminotransferase (51 U/L) concentrations.
The patient’s vital signs upon arrival at the operating room were a BP of 130/70 mmHg and SpO2 of 100%. Lead II of the ECG was monitored, and there were no unusual findings except sinus tachycardia with an HR of 100 bpm on the initial ECG. Arterial BP (ABP), central venous pressure (CVP), and esophageal temperature were monitored from the right radial artery, right subclavian vein, and lower third of the esophagus using a temperature sensor attached to the esophageal stethoscope, respectively. The initial ABP, CVP, and esophageal temperature were 140/85 mmHg, 6 mmHg, and 33.0°C, respectively. Arterial blood gas analysis was performed in the arterial line, with values calculated as for a body temperature of 37°C, which revealed the following: pH, 7.400; partial pressure of carbon dioxide (PCO2), 40 mmHg; partial pressure of oxygen (PO2), 174 mmHg; bicarbonate (HCO3), 24.8 mmol/L; base excess, 0.0; lactate, 0.17 mmol/L; ionized calcium, 1.04 mmol/L; and potassium, 3.3 mmol/L. Total intravenous (IV) anesthesia was performed using a target-controlled infusion of propofol and remifentanil in accordance with the Schnider6 and Minto7 models, respectively. Propofol and remifentanil were infused continuously with effect-site concentrations (Ce) of 1.5 µg/mL and 1.5 ng/mL, respectively. Concurrently, 50 mg of rocuronium was administered by IV bolus injection for neuromuscular blockade. Mechanical ventilation was adjusted to maintain the end-tidal carbon dioxide value at 30 to 33 mmHg.
As ABP decreased to 78/42 mmHg and CVP decreased to 4 mmHg within 10 minutes after anesthesia, the Ce of propofol and remifentanil were lowered to 0.8 µg/mL and 1.0 ng/mL, respectively, and approximately 1 L of balanced crystalloid fluid loading was performed. However, the patient’s BP did not recover, and a dopamine continuous infusion (5 µg/kg/minute) was initiated. When a surgical incision was made 10 minutes after beginning the dopamine infusion, the patient’s ABP and HR increased to 115/70 mmHg and 120 bpm, respectively, and marked ST-segment depression (−2.4 mm) was found, which was not observed on the preoperative or initial ECGs (Figure 4). Subsequently, the depth of anesthesia was increased by increasing the Ce of propofol to 1.0 to 1.2 µg/mL, followed by a continuous infusion of isosorbide dinitrate (1 µg/kg/minute). While HR was maintained at 110 to 120 bpm, ABP gradually decreased to 70/40 mmHg after approximately 5 minutes. To treat the hypotension, isosorbide dinitrate was discontinued, and a bolus injection of 100 µg phenylephrine was administered, after which, ABP increased temporarily to 95/60 mmHg. Subsequently, HR decreased suddenly and rapidly, and ventricular fibrillation occurred. The anesthesiologists immediately discontinued the surgery and performed cardiopulmonary resuscitation (CPR). The patient’s esophageal temperature at the time was 32.0°C.
Figure 4. Intraoperative electrocardiogram of the patient showing sinus tachycardia with significant ST depression. The ST depression on the monitor represented approximately −2.4 at the time.
Chest compressions were performed with periodic defibrillation with 200 J, and epinephrine injections were administered every 3 minutes; these measures elicited no response. During CPR, ECG monitoring showed temporary severe bradycardia (HR, 10–30 bpm). However, this was considered pulseless electrical activity because no pulse was detected in the patient’s femoral artery. Soon afterward, there was a rapid transition to ventricular fibrillation, and CPR was repeated. At the time, the patient’s esophageal temperature dropped to 30.0°C. According to the refractory state after the defibrillation and drug injections, we considered that the cardiac arrest was caused by hypothermia. To reverse the cause, a forced-air warming device and a warming blanket (Bair Hugger®; Arizant Healthcare Inc., Eden Prairie, MN, USA) were used for surface warming, the IV fluids were warmed, and surgical wound closure and dressing were performed rapidly to prevent surface heat loss. The first arterial blood gas analysis after ventricular fibrillation, calculated as for a body temperature of 37°C, showed the following: pH, 7.690; PCO2, 15 mmHg; PO2, 260 mmHg; HCO3, 18.1 mmol/L; base excess, −1.2; lactate, 0.33 mmol/L; ionized calcium, 0.88 mmol/L; and potassium, 5.0 mmol/L. Follow-up arterial blood gas analyses were repeated periodically. As we continued to perform CPR, the patient developed severe metabolic acidosis and hypocalcemia characterized by a pH of 6.840, HCO3 of 15.4 mmol/L, and ionized calcium of 0.66 mmol/L. To correct the abnormal findings, we administered appropriate amounts of sodium bicarbonate and calcium chloride. After an hour-long CPR and rewarming, the patient’s esophageal temperature recovered to 33.0°C, which was followed by observing sinus rhythm characterized by an HR of 75 bpm on an ECG monitor and detecting a pulse in the femoral artery with a characteristic ABP waveform that eventually led to the recovery of BP to 130/55 mmHg. Upon confirming recovery of spontaneous circulation (ROSC), the patient was transported to the ICU while we administered dopamine (10 µg/kg/minute), norepinephrine (0.1 µg/kg/minute), and vasopressin (2 U/hour). ABP, HR, and esophageal temperature during anesthesia are shown in Figure 5.
Figure 5. Changes in vital signs (blood pressure (BP), heart rate (HR), and esophageal temperature) over time (5-minute intervals).
bpm, beats per minute.
Cardiac biomarker testing performed after the patient’s transfer to the ICU showed an increased troponin-T level of 1.432 µg/L. Portable Doppler transthoracic echocardiography (TTE) revealed total occlusion of the left main coronary artery, which had undergone PCI previously. Because emergency coronary artery bypass grafting (CABG) was impossible at the time, repeated PCI was considered an alternative treatment option. However, even after his transfer to the ICU, the patient experienced repetitive ventricular fibrillation. Four hours after the patient’s transfer to the ICU, the patient’s guardian requested that CPR should be stopped. Without resuscitation, the patient died.
Discussion
In this case, the patient, who was receiving therapeutic hypothermia, underwent emergency craniectomy owing to worsening brain hemorrhage, and cardiac arrest occurred during general anesthesia. We deduced that the cardiac arrest was owing to the additional decrease in the patient’s core body temperature and myocardial infarction caused by restenosis of the previous PCI-treated lesion.
Therapeutic hypothermia is used to improve neurological outcomes in cardiac arrest, TBI, stroke, and hypoxic–ischemic encephalopathy.1,4 Studies on the efficacy of therapeutic hypothermia for TBI have been less convincing, and they show conflicting results. However, hypothermia is consistently used as an option for TBI because it reduces intracranial pressure and the cerebral metabolic rate, thereby protecting neuronal function in the brain and decreasing the wide range of injuries or reparative processes by attenuating temperature-sensitive mechanisms, such as excitotoxicity, free radical generation, apoptosis, and inflammation.1,2
The cardiovascular effects associated with hypothermia are complex. Hypothermia affects the electrical conduction of myocardial cells, which can induce hypotension or bradycardia and even increase the risk of ventricular fibrillation by prolonging repolarization.4,8 However, mild hypothermia of 32.0°C to 35.0°C demonstrates cardioprotective effects by decreasing myocardial oxygen demand and increasing myocardial perfusion through vasodilation of coronary vessels.9 Therefore, there has been a growing interest in the clinical use of mild therapeutic hypothermia as an adjunctive therapy during ischemic heart disease.10 Our patient received mild therapeutic hypothermia with a target core temperature of 33.0°C, which may have yielded cardioprotective as well as neuroprotective benefits.11
During general anesthesia, hypothermia is induced following an increase in heat loss caused by peripheral vasodilation after the injection of anesthetic drugs, heat redistribution from the core to the periphery, and impaired normal thermoregulatory responses, such as vasoconstriction or shivering.5 In our patient, the cold operating room temperature and irrigating an uncovered surgical area in the brain with cold saline also may have contributed to the additional core body temperature drop in addition to the effects of general anesthesia.12 However, in our case, neurosurgeons requested that anesthesiologists not perform warming during anesthesia. Instead, esophageal temperature and signs that could result from hypothermia, such as ECG changes, were carefully monitored. The esophageal temperature decreased from 33.0°C, which was measured immediately after entering the operation room, to 32.0°C at the time of cardiac arrest. The esophageal temperature further decreased to a minimum of 30.0°C during CPR. With rewarming, the temperature recovered to 33.0°C and finally, ROSC was achieved. Considering that chest compressions, drugs, and defibrillation did not induce ROSC, which was achieved with the recovery of an esophageal temperature of 33.0°C, the additional rapid fall in the patient’s core body temperature during anesthesia may have caused the cardiac arrest.
In this case study, we suspected that restenosis at the site of the previous stent insertion contributed to the cardiac arrest, given the following three findings: First, the patient had a normal preoperative ECG with no signs of ischemia but developed ST-segment depression during anesthesia. Second, repetitive cardiac arrests occurred even after achieving ROSC with rewarming. Third, an increased troponin-T concentration was observed, and suspected re-stenotic occlusion of the left main coronary artery was confirmed by Doppler TTE after the patient’s transfer to the ICU. Accordingly, myocardial infarction was considered the main cause of death.
The mechanisms of myocardial infarction owing to stent restenosis after DES insertion are rapid stent thrombosis or profuse in-stent restenosis (ISR).13 We assume that the lesion characteristics in the left main bifurcation of the coronary artery and DAPT interruption owing to the increasing amount of cerebral hemorrhage contributed to the stent restenosis in this patient. As evidence in support of this assumption, previous studies have shown that stenting at the left main bifurcation and DAPT interruption are strong risk factors for stent thrombosis or ISR.14,15 While controversial, not interrupting aspirin perioperatively could have avoided poor outcomes.16 However, there were no clinical signs related to expected ongoing restenosis of the coronary artery or myocardial ischemia preoperatively in our patient. We suspect that the possible ischemic signs may have been masked by the aforementioned cardioprotective benefits of therapeutic hypothermia. We also believe that the following three conditions contributed to the intraoperative myocardial ischemia and infarction caused by acute restenosis of the coronary stent: First, intraoperative hypotension may have contributed to myocardial injury. The IV anesthetics used during anesthesia, propofol and remifentanil, induced redistribution of blood volume by peripheral vasodilation, resulting in profound hypotension.4 Profound hypotension was strongly associated with myocardial injury and especially harmful for this patient, who had preexisting ischemic heart disease.17 Second, tachycardia, which may have been caused by the inotrope dopamine, used to compensate for the low BP or the stress response to surgical stimulation from shallow anesthetic depth, may have triggered myocardial ischemia by decreasing oxygen delivery and increasing oxygen consumption.17,18 If cardiogenic shock is suspected, as in our case, norepinephrine could be a better option than dopamine.19 Third, the bolus injection of phenylephrine, a selective α1-adrenergic agonist, just before cardiac arrest may have temporarily and abruptly increased cardiac afterload and impaired myocardial perfusion.20
Had we known that coronary artery occlusion would result from hypothermia-induced cardiac arrest and coronary stent restenosis, the treatment of choice would have been CABG surgery while maintaining circulation with emergency cardiopulmonary bypass (CPB) or extracorporeal membrane oxygenation (ECMO) via the femoral artery and vein.11 However, it was difficult to determine the etiology of the cardiac arrest at the time. Even if we had known the cause, ideal treatment would have been impossible, given circumstances beyond our control; all thoracic surgeons were busy operating, and emergency CPB and ECMO were not available. As an alternative, repeat PCI after the patient achieved ROSC could have been performed. However, the patient had a prolonged CPR duration, and he did not maintain the ROSC state for a period sufficient for intervention, which led to a poor outcome.
In this case, it was not possible to accurately determine the reason for the cardiac arrest and for not achieving consistent ROSC. Instead, we conclude that cardiac arrest developed and persisted as a result of the unintentional severe hypothermia that occurred during general anesthesia and/or coronary artery restenosis at the site of the previous PCI. A thorough evaluation of heart-related history is required for TBI patients who have a history of ischemic heart diseases and who undergo therapeutic hypothermia because there may be undiscovered coronary insufficiency owing to the cardioprotective effects of therapeutic hypothermia. In addition, careful monitoring of the patient’s core body temperature, and preventive and countermeasures should be fully prepared and taken against adverse hemodynamic effects, even cardiac arrest, which can occur because of additive hypothermic responses during anesthesia.
Ethics statement: This case report was approved by our Institutional Review Board (EMC 2020-05-005-001), and the case report was performed in compliance with the EQUATOR Network guidelines (the CARE guidelines). All patient details were de-identified, and written consent was obtained from the patient’s guardian for treatment and publication of this report.
Declaration of conflicting interest: The authors declare that there is no conflict of interest.
Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
ORCID iDs: Dong Ho Park https://orcid.org/0000-0002-6587-3756
Tae Woo Kim https://orcid.org/0000-0002-2115-6448
Mo Se Kim https://orcid.org/0000-0001-7362-0030 | 50 mg (milligrams). | DrugDosage | CC BY-NC | 33499678 | 19,395,621 | 2021-01 |
What was the outcome of reaction 'Bradycardia'? | Cardiac arrest caused by accidental severe hypothermia and myocardial infarction during general anesthesia.
Therapeutic hypothermia is often used for traumatic brain injury because of its neuroprotective effect and decreased secondary brain injury. However, this procedure lacks clinical evidence supporting its efficacy, and adverse outcomes have been reported during general anesthesia. A 61-year-old man with a history of percutaneous coronary intervention (PCI) was admitted with traumatic brain injury. Immediately after admission, he underwent mild therapeutic hypothermia with a target temperature of 33.0°C for neuroprotection. During general anesthesia for emergency surgery because he developed a mass effect, hypothermic cardiac arrest occurred following an additional decrease in the core body temperature. Moreover, myocardial infarction caused by restenosis of the previous PCI lesion also contributed to the cardiac arrest. Although the patient recovered spontaneous circulation after an hour-long cardiopulmonary resuscitation with rewarming, he eventually died of subsequent repetitive cardiac arrests. When anesthetizing patients undergoing therapeutic hypothermia, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.
Introduction
Therapeutic hypothermia is an attractive strategy for traumatic brain injury (TBI), which demonstrates neuroprotective effects and decreases secondary brain injury.1 Despite the lack of clinical evidence regarding efficacy, therapeutic hypothermia after TBI has been used for at least half a century because of its potential benefits, which are supported by numerous preclinical studies.2,3 However, there have been no reports on the danger of unintentional severe hypothermia during general anesthesia caused by the additional drop in temperature of patients undergoing therapeutic hypothermia. When anesthetizing patients with comorbidities, such as ischemic heart disease, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.4,5
Case report
A 61-year-old man (height, 168 cm; weight, 70 kg) presented to our emergency room (ER) with head trauma after falling on a cement floor from a 5-meter-tall construction site. Upon arrival at the ER, the patient had a Glasgow coma scale score of 11 and the following vital signs: blood pressure (BP), 147/83 mmHg; heart rate (HR), 86 beats per minute (bpm); respiratory rate, 24 breaths per minute; body temperature from the auditory canal, 36.8°C; and percutaneous oxygen saturation (SpO2), 97%. Upon discovering signs of acute subdural hemorrhage, subarachnoid hemorrhage, and intraventricular hemorrhage from a brain computed tomography (CT) scan performed in the ER, we admitted the patient to the intensive care unit (ICU) for close monitoring. After approximately 6 hours, we performed follow-up brain CT because the patient’s mental status had deteriorated from drowsy to stupor. Upon observing a mass effect characterized by an increase in the hemorrhage volume and a left midline shift on the CT images, the patient was scheduled for immediate intubation and emergency decompressive craniectomy (Figure 1).
Figure 1. Brain computed tomography (CT) images taken 6 hours after intensive care unit admission. (a) Axial view showing acute subarachnoid hemorrhage in the peri-mesencephalic cistern (arrow A) and subdural hemorrhage in the right inferior temporal (arrow B) and right parietal (arrow C) regions. (b) Axial view showing acute subdural hemorrhage in the right temporal region (black arrow) and left midline shift.
Although the patient had a history of ischemic heart disease with percutaneous coronary intervention (PCI), it was difficult to obtain accurate information because there was no relevant data in the patient’s medical records from the authors’ hospital, and his drowsy state interfered with detailed interviews. According to the chest CT scan performed immediately after admission, coronary stents had been inserted at the bifurcation site of the left main coronary artery (Figure 2). It also appeared that the inserted stents were drug-eluting stents (DES) considering that he had received dual antiplatelet therapy (DAPT) with aspirin and clopidogrel (Plavix®; Sanofi, Bridgewater, NJ, USA) a switching platelet P2Y12 inhibitor, for approximately 4 months. However, DAPT was discontinued owing to the worsening cerebral hemorrhage after hospitalization; bridging antiplatelet therapy was also not performed. Before the surgery, the patient showed a normal electrocardiogram (ECG), and cardiac biomarker testing revealed a troponin-T concentration of 0.006 µg/L, which was within normal limits (Figure 3). The patient demonstrated satisfactory functional capacity to be able to work at a construction site just before the trauma. Therefore, an additional preoperative cardiac evaluation was not considered necessary.
Figure 2. Chest computed tomography (CT) images showing coronary stents in the main bifurcation of the left coronary arteries (white arrow in (a) and (b)).
Figure 3. Preoperative electrocardiogram of the patient showing normal sinus rhythm with a heart rate of 95 bpm.
bpm, beats per minute.
Immediately after admission to the ICU, the patient’s bladder temperature was 38.0°C. Using a hydrogel-coated water-circulating pad (Arctic Sun®; Medivance Inc., Louisville, CO, USA), which is a surface cooling device, the patient underwent mild therapeutic hypothermia with a target temperature of 33.0°C. During therapeutic hypothermia, sedation and shivering control were performed using continuous rate infusions of midazolam at 5 mg/hour and remifentanil at 0.2 µg/kg/minute. The patient’s bladder temperature was 33.4°C just before he was transferred to the operating room. Preoperative laboratory tests showed no unusual findings except for mildly elevated aspartate aminotransferase (73 U/L) and alanine aminotransferase (51 U/L) concentrations.
The patient’s vital signs upon arrival at the operating room were a BP of 130/70 mmHg and SpO2 of 100%. Lead II of the ECG was monitored, and there were no unusual findings except sinus tachycardia with an HR of 100 bpm on the initial ECG. Arterial BP (ABP), central venous pressure (CVP), and esophageal temperature were monitored from the right radial artery, right subclavian vein, and lower third of the esophagus using a temperature sensor attached to the esophageal stethoscope, respectively. The initial ABP, CVP, and esophageal temperature were 140/85 mmHg, 6 mmHg, and 33.0°C, respectively. Arterial blood gas analysis was performed in the arterial line, with values calculated as for a body temperature of 37°C, which revealed the following: pH, 7.400; partial pressure of carbon dioxide (PCO2), 40 mmHg; partial pressure of oxygen (PO2), 174 mmHg; bicarbonate (HCO3), 24.8 mmol/L; base excess, 0.0; lactate, 0.17 mmol/L; ionized calcium, 1.04 mmol/L; and potassium, 3.3 mmol/L. Total intravenous (IV) anesthesia was performed using a target-controlled infusion of propofol and remifentanil in accordance with the Schnider6 and Minto7 models, respectively. Propofol and remifentanil were infused continuously with effect-site concentrations (Ce) of 1.5 µg/mL and 1.5 ng/mL, respectively. Concurrently, 50 mg of rocuronium was administered by IV bolus injection for neuromuscular blockade. Mechanical ventilation was adjusted to maintain the end-tidal carbon dioxide value at 30 to 33 mmHg.
As ABP decreased to 78/42 mmHg and CVP decreased to 4 mmHg within 10 minutes after anesthesia, the Ce of propofol and remifentanil were lowered to 0.8 µg/mL and 1.0 ng/mL, respectively, and approximately 1 L of balanced crystalloid fluid loading was performed. However, the patient’s BP did not recover, and a dopamine continuous infusion (5 µg/kg/minute) was initiated. When a surgical incision was made 10 minutes after beginning the dopamine infusion, the patient’s ABP and HR increased to 115/70 mmHg and 120 bpm, respectively, and marked ST-segment depression (−2.4 mm) was found, which was not observed on the preoperative or initial ECGs (Figure 4). Subsequently, the depth of anesthesia was increased by increasing the Ce of propofol to 1.0 to 1.2 µg/mL, followed by a continuous infusion of isosorbide dinitrate (1 µg/kg/minute). While HR was maintained at 110 to 120 bpm, ABP gradually decreased to 70/40 mmHg after approximately 5 minutes. To treat the hypotension, isosorbide dinitrate was discontinued, and a bolus injection of 100 µg phenylephrine was administered, after which, ABP increased temporarily to 95/60 mmHg. Subsequently, HR decreased suddenly and rapidly, and ventricular fibrillation occurred. The anesthesiologists immediately discontinued the surgery and performed cardiopulmonary resuscitation (CPR). The patient’s esophageal temperature at the time was 32.0°C.
Figure 4. Intraoperative electrocardiogram of the patient showing sinus tachycardia with significant ST depression. The ST depression on the monitor represented approximately −2.4 at the time.
Chest compressions were performed with periodic defibrillation with 200 J, and epinephrine injections were administered every 3 minutes; these measures elicited no response. During CPR, ECG monitoring showed temporary severe bradycardia (HR, 10–30 bpm). However, this was considered pulseless electrical activity because no pulse was detected in the patient’s femoral artery. Soon afterward, there was a rapid transition to ventricular fibrillation, and CPR was repeated. At the time, the patient’s esophageal temperature dropped to 30.0°C. According to the refractory state after the defibrillation and drug injections, we considered that the cardiac arrest was caused by hypothermia. To reverse the cause, a forced-air warming device and a warming blanket (Bair Hugger®; Arizant Healthcare Inc., Eden Prairie, MN, USA) were used for surface warming, the IV fluids were warmed, and surgical wound closure and dressing were performed rapidly to prevent surface heat loss. The first arterial blood gas analysis after ventricular fibrillation, calculated as for a body temperature of 37°C, showed the following: pH, 7.690; PCO2, 15 mmHg; PO2, 260 mmHg; HCO3, 18.1 mmol/L; base excess, −1.2; lactate, 0.33 mmol/L; ionized calcium, 0.88 mmol/L; and potassium, 5.0 mmol/L. Follow-up arterial blood gas analyses were repeated periodically. As we continued to perform CPR, the patient developed severe metabolic acidosis and hypocalcemia characterized by a pH of 6.840, HCO3 of 15.4 mmol/L, and ionized calcium of 0.66 mmol/L. To correct the abnormal findings, we administered appropriate amounts of sodium bicarbonate and calcium chloride. After an hour-long CPR and rewarming, the patient’s esophageal temperature recovered to 33.0°C, which was followed by observing sinus rhythm characterized by an HR of 75 bpm on an ECG monitor and detecting a pulse in the femoral artery with a characteristic ABP waveform that eventually led to the recovery of BP to 130/55 mmHg. Upon confirming recovery of spontaneous circulation (ROSC), the patient was transported to the ICU while we administered dopamine (10 µg/kg/minute), norepinephrine (0.1 µg/kg/minute), and vasopressin (2 U/hour). ABP, HR, and esophageal temperature during anesthesia are shown in Figure 5.
Figure 5. Changes in vital signs (blood pressure (BP), heart rate (HR), and esophageal temperature) over time (5-minute intervals).
bpm, beats per minute.
Cardiac biomarker testing performed after the patient’s transfer to the ICU showed an increased troponin-T level of 1.432 µg/L. Portable Doppler transthoracic echocardiography (TTE) revealed total occlusion of the left main coronary artery, which had undergone PCI previously. Because emergency coronary artery bypass grafting (CABG) was impossible at the time, repeated PCI was considered an alternative treatment option. However, even after his transfer to the ICU, the patient experienced repetitive ventricular fibrillation. Four hours after the patient’s transfer to the ICU, the patient’s guardian requested that CPR should be stopped. Without resuscitation, the patient died.
Discussion
In this case, the patient, who was receiving therapeutic hypothermia, underwent emergency craniectomy owing to worsening brain hemorrhage, and cardiac arrest occurred during general anesthesia. We deduced that the cardiac arrest was owing to the additional decrease in the patient’s core body temperature and myocardial infarction caused by restenosis of the previous PCI-treated lesion.
Therapeutic hypothermia is used to improve neurological outcomes in cardiac arrest, TBI, stroke, and hypoxic–ischemic encephalopathy.1,4 Studies on the efficacy of therapeutic hypothermia for TBI have been less convincing, and they show conflicting results. However, hypothermia is consistently used as an option for TBI because it reduces intracranial pressure and the cerebral metabolic rate, thereby protecting neuronal function in the brain and decreasing the wide range of injuries or reparative processes by attenuating temperature-sensitive mechanisms, such as excitotoxicity, free radical generation, apoptosis, and inflammation.1,2
The cardiovascular effects associated with hypothermia are complex. Hypothermia affects the electrical conduction of myocardial cells, which can induce hypotension or bradycardia and even increase the risk of ventricular fibrillation by prolonging repolarization.4,8 However, mild hypothermia of 32.0°C to 35.0°C demonstrates cardioprotective effects by decreasing myocardial oxygen demand and increasing myocardial perfusion through vasodilation of coronary vessels.9 Therefore, there has been a growing interest in the clinical use of mild therapeutic hypothermia as an adjunctive therapy during ischemic heart disease.10 Our patient received mild therapeutic hypothermia with a target core temperature of 33.0°C, which may have yielded cardioprotective as well as neuroprotective benefits.11
During general anesthesia, hypothermia is induced following an increase in heat loss caused by peripheral vasodilation after the injection of anesthetic drugs, heat redistribution from the core to the periphery, and impaired normal thermoregulatory responses, such as vasoconstriction or shivering.5 In our patient, the cold operating room temperature and irrigating an uncovered surgical area in the brain with cold saline also may have contributed to the additional core body temperature drop in addition to the effects of general anesthesia.12 However, in our case, neurosurgeons requested that anesthesiologists not perform warming during anesthesia. Instead, esophageal temperature and signs that could result from hypothermia, such as ECG changes, were carefully monitored. The esophageal temperature decreased from 33.0°C, which was measured immediately after entering the operation room, to 32.0°C at the time of cardiac arrest. The esophageal temperature further decreased to a minimum of 30.0°C during CPR. With rewarming, the temperature recovered to 33.0°C and finally, ROSC was achieved. Considering that chest compressions, drugs, and defibrillation did not induce ROSC, which was achieved with the recovery of an esophageal temperature of 33.0°C, the additional rapid fall in the patient’s core body temperature during anesthesia may have caused the cardiac arrest.
In this case study, we suspected that restenosis at the site of the previous stent insertion contributed to the cardiac arrest, given the following three findings: First, the patient had a normal preoperative ECG with no signs of ischemia but developed ST-segment depression during anesthesia. Second, repetitive cardiac arrests occurred even after achieving ROSC with rewarming. Third, an increased troponin-T concentration was observed, and suspected re-stenotic occlusion of the left main coronary artery was confirmed by Doppler TTE after the patient’s transfer to the ICU. Accordingly, myocardial infarction was considered the main cause of death.
The mechanisms of myocardial infarction owing to stent restenosis after DES insertion are rapid stent thrombosis or profuse in-stent restenosis (ISR).13 We assume that the lesion characteristics in the left main bifurcation of the coronary artery and DAPT interruption owing to the increasing amount of cerebral hemorrhage contributed to the stent restenosis in this patient. As evidence in support of this assumption, previous studies have shown that stenting at the left main bifurcation and DAPT interruption are strong risk factors for stent thrombosis or ISR.14,15 While controversial, not interrupting aspirin perioperatively could have avoided poor outcomes.16 However, there were no clinical signs related to expected ongoing restenosis of the coronary artery or myocardial ischemia preoperatively in our patient. We suspect that the possible ischemic signs may have been masked by the aforementioned cardioprotective benefits of therapeutic hypothermia. We also believe that the following three conditions contributed to the intraoperative myocardial ischemia and infarction caused by acute restenosis of the coronary stent: First, intraoperative hypotension may have contributed to myocardial injury. The IV anesthetics used during anesthesia, propofol and remifentanil, induced redistribution of blood volume by peripheral vasodilation, resulting in profound hypotension.4 Profound hypotension was strongly associated with myocardial injury and especially harmful for this patient, who had preexisting ischemic heart disease.17 Second, tachycardia, which may have been caused by the inotrope dopamine, used to compensate for the low BP or the stress response to surgical stimulation from shallow anesthetic depth, may have triggered myocardial ischemia by decreasing oxygen delivery and increasing oxygen consumption.17,18 If cardiogenic shock is suspected, as in our case, norepinephrine could be a better option than dopamine.19 Third, the bolus injection of phenylephrine, a selective α1-adrenergic agonist, just before cardiac arrest may have temporarily and abruptly increased cardiac afterload and impaired myocardial perfusion.20
Had we known that coronary artery occlusion would result from hypothermia-induced cardiac arrest and coronary stent restenosis, the treatment of choice would have been CABG surgery while maintaining circulation with emergency cardiopulmonary bypass (CPB) or extracorporeal membrane oxygenation (ECMO) via the femoral artery and vein.11 However, it was difficult to determine the etiology of the cardiac arrest at the time. Even if we had known the cause, ideal treatment would have been impossible, given circumstances beyond our control; all thoracic surgeons were busy operating, and emergency CPB and ECMO were not available. As an alternative, repeat PCI after the patient achieved ROSC could have been performed. However, the patient had a prolonged CPR duration, and he did not maintain the ROSC state for a period sufficient for intervention, which led to a poor outcome.
In this case, it was not possible to accurately determine the reason for the cardiac arrest and for not achieving consistent ROSC. Instead, we conclude that cardiac arrest developed and persisted as a result of the unintentional severe hypothermia that occurred during general anesthesia and/or coronary artery restenosis at the site of the previous PCI. A thorough evaluation of heart-related history is required for TBI patients who have a history of ischemic heart diseases and who undergo therapeutic hypothermia because there may be undiscovered coronary insufficiency owing to the cardioprotective effects of therapeutic hypothermia. In addition, careful monitoring of the patient’s core body temperature, and preventive and countermeasures should be fully prepared and taken against adverse hemodynamic effects, even cardiac arrest, which can occur because of additive hypothermic responses during anesthesia.
Ethics statement: This case report was approved by our Institutional Review Board (EMC 2020-05-005-001), and the case report was performed in compliance with the EQUATOR Network guidelines (the CARE guidelines). All patient details were de-identified, and written consent was obtained from the patient’s guardian for treatment and publication of this report.
Declaration of conflicting interest: The authors declare that there is no conflict of interest.
Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
ORCID iDs: Dong Ho Park https://orcid.org/0000-0002-6587-3756
Tae Woo Kim https://orcid.org/0000-0002-2115-6448
Mo Se Kim https://orcid.org/0000-0001-7362-0030 | Recovered | ReactionOutcome | CC BY-NC | 33499678 | 19,429,456 | 2021-01 |
What was the outcome of reaction 'Cardiac arrest'? | Cardiac arrest caused by accidental severe hypothermia and myocardial infarction during general anesthesia.
Therapeutic hypothermia is often used for traumatic brain injury because of its neuroprotective effect and decreased secondary brain injury. However, this procedure lacks clinical evidence supporting its efficacy, and adverse outcomes have been reported during general anesthesia. A 61-year-old man with a history of percutaneous coronary intervention (PCI) was admitted with traumatic brain injury. Immediately after admission, he underwent mild therapeutic hypothermia with a target temperature of 33.0°C for neuroprotection. During general anesthesia for emergency surgery because he developed a mass effect, hypothermic cardiac arrest occurred following an additional decrease in the core body temperature. Moreover, myocardial infarction caused by restenosis of the previous PCI lesion also contributed to the cardiac arrest. Although the patient recovered spontaneous circulation after an hour-long cardiopulmonary resuscitation with rewarming, he eventually died of subsequent repetitive cardiac arrests. When anesthetizing patients undergoing therapeutic hypothermia, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.
Introduction
Therapeutic hypothermia is an attractive strategy for traumatic brain injury (TBI), which demonstrates neuroprotective effects and decreases secondary brain injury.1 Despite the lack of clinical evidence regarding efficacy, therapeutic hypothermia after TBI has been used for at least half a century because of its potential benefits, which are supported by numerous preclinical studies.2,3 However, there have been no reports on the danger of unintentional severe hypothermia during general anesthesia caused by the additional drop in temperature of patients undergoing therapeutic hypothermia. When anesthetizing patients with comorbidities, such as ischemic heart disease, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.4,5
Case report
A 61-year-old man (height, 168 cm; weight, 70 kg) presented to our emergency room (ER) with head trauma after falling on a cement floor from a 5-meter-tall construction site. Upon arrival at the ER, the patient had a Glasgow coma scale score of 11 and the following vital signs: blood pressure (BP), 147/83 mmHg; heart rate (HR), 86 beats per minute (bpm); respiratory rate, 24 breaths per minute; body temperature from the auditory canal, 36.8°C; and percutaneous oxygen saturation (SpO2), 97%. Upon discovering signs of acute subdural hemorrhage, subarachnoid hemorrhage, and intraventricular hemorrhage from a brain computed tomography (CT) scan performed in the ER, we admitted the patient to the intensive care unit (ICU) for close monitoring. After approximately 6 hours, we performed follow-up brain CT because the patient’s mental status had deteriorated from drowsy to stupor. Upon observing a mass effect characterized by an increase in the hemorrhage volume and a left midline shift on the CT images, the patient was scheduled for immediate intubation and emergency decompressive craniectomy (Figure 1).
Figure 1. Brain computed tomography (CT) images taken 6 hours after intensive care unit admission. (a) Axial view showing acute subarachnoid hemorrhage in the peri-mesencephalic cistern (arrow A) and subdural hemorrhage in the right inferior temporal (arrow B) and right parietal (arrow C) regions. (b) Axial view showing acute subdural hemorrhage in the right temporal region (black arrow) and left midline shift.
Although the patient had a history of ischemic heart disease with percutaneous coronary intervention (PCI), it was difficult to obtain accurate information because there was no relevant data in the patient’s medical records from the authors’ hospital, and his drowsy state interfered with detailed interviews. According to the chest CT scan performed immediately after admission, coronary stents had been inserted at the bifurcation site of the left main coronary artery (Figure 2). It also appeared that the inserted stents were drug-eluting stents (DES) considering that he had received dual antiplatelet therapy (DAPT) with aspirin and clopidogrel (Plavix®; Sanofi, Bridgewater, NJ, USA) a switching platelet P2Y12 inhibitor, for approximately 4 months. However, DAPT was discontinued owing to the worsening cerebral hemorrhage after hospitalization; bridging antiplatelet therapy was also not performed. Before the surgery, the patient showed a normal electrocardiogram (ECG), and cardiac biomarker testing revealed a troponin-T concentration of 0.006 µg/L, which was within normal limits (Figure 3). The patient demonstrated satisfactory functional capacity to be able to work at a construction site just before the trauma. Therefore, an additional preoperative cardiac evaluation was not considered necessary.
Figure 2. Chest computed tomography (CT) images showing coronary stents in the main bifurcation of the left coronary arteries (white arrow in (a) and (b)).
Figure 3. Preoperative electrocardiogram of the patient showing normal sinus rhythm with a heart rate of 95 bpm.
bpm, beats per minute.
Immediately after admission to the ICU, the patient’s bladder temperature was 38.0°C. Using a hydrogel-coated water-circulating pad (Arctic Sun®; Medivance Inc., Louisville, CO, USA), which is a surface cooling device, the patient underwent mild therapeutic hypothermia with a target temperature of 33.0°C. During therapeutic hypothermia, sedation and shivering control were performed using continuous rate infusions of midazolam at 5 mg/hour and remifentanil at 0.2 µg/kg/minute. The patient’s bladder temperature was 33.4°C just before he was transferred to the operating room. Preoperative laboratory tests showed no unusual findings except for mildly elevated aspartate aminotransferase (73 U/L) and alanine aminotransferase (51 U/L) concentrations.
The patient’s vital signs upon arrival at the operating room were a BP of 130/70 mmHg and SpO2 of 100%. Lead II of the ECG was monitored, and there were no unusual findings except sinus tachycardia with an HR of 100 bpm on the initial ECG. Arterial BP (ABP), central venous pressure (CVP), and esophageal temperature were monitored from the right radial artery, right subclavian vein, and lower third of the esophagus using a temperature sensor attached to the esophageal stethoscope, respectively. The initial ABP, CVP, and esophageal temperature were 140/85 mmHg, 6 mmHg, and 33.0°C, respectively. Arterial blood gas analysis was performed in the arterial line, with values calculated as for a body temperature of 37°C, which revealed the following: pH, 7.400; partial pressure of carbon dioxide (PCO2), 40 mmHg; partial pressure of oxygen (PO2), 174 mmHg; bicarbonate (HCO3), 24.8 mmol/L; base excess, 0.0; lactate, 0.17 mmol/L; ionized calcium, 1.04 mmol/L; and potassium, 3.3 mmol/L. Total intravenous (IV) anesthesia was performed using a target-controlled infusion of propofol and remifentanil in accordance with the Schnider6 and Minto7 models, respectively. Propofol and remifentanil were infused continuously with effect-site concentrations (Ce) of 1.5 µg/mL and 1.5 ng/mL, respectively. Concurrently, 50 mg of rocuronium was administered by IV bolus injection for neuromuscular blockade. Mechanical ventilation was adjusted to maintain the end-tidal carbon dioxide value at 30 to 33 mmHg.
As ABP decreased to 78/42 mmHg and CVP decreased to 4 mmHg within 10 minutes after anesthesia, the Ce of propofol and remifentanil were lowered to 0.8 µg/mL and 1.0 ng/mL, respectively, and approximately 1 L of balanced crystalloid fluid loading was performed. However, the patient’s BP did not recover, and a dopamine continuous infusion (5 µg/kg/minute) was initiated. When a surgical incision was made 10 minutes after beginning the dopamine infusion, the patient’s ABP and HR increased to 115/70 mmHg and 120 bpm, respectively, and marked ST-segment depression (−2.4 mm) was found, which was not observed on the preoperative or initial ECGs (Figure 4). Subsequently, the depth of anesthesia was increased by increasing the Ce of propofol to 1.0 to 1.2 µg/mL, followed by a continuous infusion of isosorbide dinitrate (1 µg/kg/minute). While HR was maintained at 110 to 120 bpm, ABP gradually decreased to 70/40 mmHg after approximately 5 minutes. To treat the hypotension, isosorbide dinitrate was discontinued, and a bolus injection of 100 µg phenylephrine was administered, after which, ABP increased temporarily to 95/60 mmHg. Subsequently, HR decreased suddenly and rapidly, and ventricular fibrillation occurred. The anesthesiologists immediately discontinued the surgery and performed cardiopulmonary resuscitation (CPR). The patient’s esophageal temperature at the time was 32.0°C.
Figure 4. Intraoperative electrocardiogram of the patient showing sinus tachycardia with significant ST depression. The ST depression on the monitor represented approximately −2.4 at the time.
Chest compressions were performed with periodic defibrillation with 200 J, and epinephrine injections were administered every 3 minutes; these measures elicited no response. During CPR, ECG monitoring showed temporary severe bradycardia (HR, 10–30 bpm). However, this was considered pulseless electrical activity because no pulse was detected in the patient’s femoral artery. Soon afterward, there was a rapid transition to ventricular fibrillation, and CPR was repeated. At the time, the patient’s esophageal temperature dropped to 30.0°C. According to the refractory state after the defibrillation and drug injections, we considered that the cardiac arrest was caused by hypothermia. To reverse the cause, a forced-air warming device and a warming blanket (Bair Hugger®; Arizant Healthcare Inc., Eden Prairie, MN, USA) were used for surface warming, the IV fluids were warmed, and surgical wound closure and dressing were performed rapidly to prevent surface heat loss. The first arterial blood gas analysis after ventricular fibrillation, calculated as for a body temperature of 37°C, showed the following: pH, 7.690; PCO2, 15 mmHg; PO2, 260 mmHg; HCO3, 18.1 mmol/L; base excess, −1.2; lactate, 0.33 mmol/L; ionized calcium, 0.88 mmol/L; and potassium, 5.0 mmol/L. Follow-up arterial blood gas analyses were repeated periodically. As we continued to perform CPR, the patient developed severe metabolic acidosis and hypocalcemia characterized by a pH of 6.840, HCO3 of 15.4 mmol/L, and ionized calcium of 0.66 mmol/L. To correct the abnormal findings, we administered appropriate amounts of sodium bicarbonate and calcium chloride. After an hour-long CPR and rewarming, the patient’s esophageal temperature recovered to 33.0°C, which was followed by observing sinus rhythm characterized by an HR of 75 bpm on an ECG monitor and detecting a pulse in the femoral artery with a characteristic ABP waveform that eventually led to the recovery of BP to 130/55 mmHg. Upon confirming recovery of spontaneous circulation (ROSC), the patient was transported to the ICU while we administered dopamine (10 µg/kg/minute), norepinephrine (0.1 µg/kg/minute), and vasopressin (2 U/hour). ABP, HR, and esophageal temperature during anesthesia are shown in Figure 5.
Figure 5. Changes in vital signs (blood pressure (BP), heart rate (HR), and esophageal temperature) over time (5-minute intervals).
bpm, beats per minute.
Cardiac biomarker testing performed after the patient’s transfer to the ICU showed an increased troponin-T level of 1.432 µg/L. Portable Doppler transthoracic echocardiography (TTE) revealed total occlusion of the left main coronary artery, which had undergone PCI previously. Because emergency coronary artery bypass grafting (CABG) was impossible at the time, repeated PCI was considered an alternative treatment option. However, even after his transfer to the ICU, the patient experienced repetitive ventricular fibrillation. Four hours after the patient’s transfer to the ICU, the patient’s guardian requested that CPR should be stopped. Without resuscitation, the patient died.
Discussion
In this case, the patient, who was receiving therapeutic hypothermia, underwent emergency craniectomy owing to worsening brain hemorrhage, and cardiac arrest occurred during general anesthesia. We deduced that the cardiac arrest was owing to the additional decrease in the patient’s core body temperature and myocardial infarction caused by restenosis of the previous PCI-treated lesion.
Therapeutic hypothermia is used to improve neurological outcomes in cardiac arrest, TBI, stroke, and hypoxic–ischemic encephalopathy.1,4 Studies on the efficacy of therapeutic hypothermia for TBI have been less convincing, and they show conflicting results. However, hypothermia is consistently used as an option for TBI because it reduces intracranial pressure and the cerebral metabolic rate, thereby protecting neuronal function in the brain and decreasing the wide range of injuries or reparative processes by attenuating temperature-sensitive mechanisms, such as excitotoxicity, free radical generation, apoptosis, and inflammation.1,2
The cardiovascular effects associated with hypothermia are complex. Hypothermia affects the electrical conduction of myocardial cells, which can induce hypotension or bradycardia and even increase the risk of ventricular fibrillation by prolonging repolarization.4,8 However, mild hypothermia of 32.0°C to 35.0°C demonstrates cardioprotective effects by decreasing myocardial oxygen demand and increasing myocardial perfusion through vasodilation of coronary vessels.9 Therefore, there has been a growing interest in the clinical use of mild therapeutic hypothermia as an adjunctive therapy during ischemic heart disease.10 Our patient received mild therapeutic hypothermia with a target core temperature of 33.0°C, which may have yielded cardioprotective as well as neuroprotective benefits.11
During general anesthesia, hypothermia is induced following an increase in heat loss caused by peripheral vasodilation after the injection of anesthetic drugs, heat redistribution from the core to the periphery, and impaired normal thermoregulatory responses, such as vasoconstriction or shivering.5 In our patient, the cold operating room temperature and irrigating an uncovered surgical area in the brain with cold saline also may have contributed to the additional core body temperature drop in addition to the effects of general anesthesia.12 However, in our case, neurosurgeons requested that anesthesiologists not perform warming during anesthesia. Instead, esophageal temperature and signs that could result from hypothermia, such as ECG changes, were carefully monitored. The esophageal temperature decreased from 33.0°C, which was measured immediately after entering the operation room, to 32.0°C at the time of cardiac arrest. The esophageal temperature further decreased to a minimum of 30.0°C during CPR. With rewarming, the temperature recovered to 33.0°C and finally, ROSC was achieved. Considering that chest compressions, drugs, and defibrillation did not induce ROSC, which was achieved with the recovery of an esophageal temperature of 33.0°C, the additional rapid fall in the patient’s core body temperature during anesthesia may have caused the cardiac arrest.
In this case study, we suspected that restenosis at the site of the previous stent insertion contributed to the cardiac arrest, given the following three findings: First, the patient had a normal preoperative ECG with no signs of ischemia but developed ST-segment depression during anesthesia. Second, repetitive cardiac arrests occurred even after achieving ROSC with rewarming. Third, an increased troponin-T concentration was observed, and suspected re-stenotic occlusion of the left main coronary artery was confirmed by Doppler TTE after the patient’s transfer to the ICU. Accordingly, myocardial infarction was considered the main cause of death.
The mechanisms of myocardial infarction owing to stent restenosis after DES insertion are rapid stent thrombosis or profuse in-stent restenosis (ISR).13 We assume that the lesion characteristics in the left main bifurcation of the coronary artery and DAPT interruption owing to the increasing amount of cerebral hemorrhage contributed to the stent restenosis in this patient. As evidence in support of this assumption, previous studies have shown that stenting at the left main bifurcation and DAPT interruption are strong risk factors for stent thrombosis or ISR.14,15 While controversial, not interrupting aspirin perioperatively could have avoided poor outcomes.16 However, there were no clinical signs related to expected ongoing restenosis of the coronary artery or myocardial ischemia preoperatively in our patient. We suspect that the possible ischemic signs may have been masked by the aforementioned cardioprotective benefits of therapeutic hypothermia. We also believe that the following three conditions contributed to the intraoperative myocardial ischemia and infarction caused by acute restenosis of the coronary stent: First, intraoperative hypotension may have contributed to myocardial injury. The IV anesthetics used during anesthesia, propofol and remifentanil, induced redistribution of blood volume by peripheral vasodilation, resulting in profound hypotension.4 Profound hypotension was strongly associated with myocardial injury and especially harmful for this patient, who had preexisting ischemic heart disease.17 Second, tachycardia, which may have been caused by the inotrope dopamine, used to compensate for the low BP or the stress response to surgical stimulation from shallow anesthetic depth, may have triggered myocardial ischemia by decreasing oxygen delivery and increasing oxygen consumption.17,18 If cardiogenic shock is suspected, as in our case, norepinephrine could be a better option than dopamine.19 Third, the bolus injection of phenylephrine, a selective α1-adrenergic agonist, just before cardiac arrest may have temporarily and abruptly increased cardiac afterload and impaired myocardial perfusion.20
Had we known that coronary artery occlusion would result from hypothermia-induced cardiac arrest and coronary stent restenosis, the treatment of choice would have been CABG surgery while maintaining circulation with emergency cardiopulmonary bypass (CPB) or extracorporeal membrane oxygenation (ECMO) via the femoral artery and vein.11 However, it was difficult to determine the etiology of the cardiac arrest at the time. Even if we had known the cause, ideal treatment would have been impossible, given circumstances beyond our control; all thoracic surgeons were busy operating, and emergency CPB and ECMO were not available. As an alternative, repeat PCI after the patient achieved ROSC could have been performed. However, the patient had a prolonged CPR duration, and he did not maintain the ROSC state for a period sufficient for intervention, which led to a poor outcome.
In this case, it was not possible to accurately determine the reason for the cardiac arrest and for not achieving consistent ROSC. Instead, we conclude that cardiac arrest developed and persisted as a result of the unintentional severe hypothermia that occurred during general anesthesia and/or coronary artery restenosis at the site of the previous PCI. A thorough evaluation of heart-related history is required for TBI patients who have a history of ischemic heart diseases and who undergo therapeutic hypothermia because there may be undiscovered coronary insufficiency owing to the cardioprotective effects of therapeutic hypothermia. In addition, careful monitoring of the patient’s core body temperature, and preventive and countermeasures should be fully prepared and taken against adverse hemodynamic effects, even cardiac arrest, which can occur because of additive hypothermic responses during anesthesia.
Ethics statement: This case report was approved by our Institutional Review Board (EMC 2020-05-005-001), and the case report was performed in compliance with the EQUATOR Network guidelines (the CARE guidelines). All patient details were de-identified, and written consent was obtained from the patient’s guardian for treatment and publication of this report.
Declaration of conflicting interest: The authors declare that there is no conflict of interest.
Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
ORCID iDs: Dong Ho Park https://orcid.org/0000-0002-6587-3756
Tae Woo Kim https://orcid.org/0000-0002-2115-6448
Mo Se Kim https://orcid.org/0000-0001-7362-0030 | Fatal | ReactionOutcome | CC BY-NC | 33499678 | 19,429,456 | 2021-01 |
What was the outcome of reaction 'Hypotension'? | Cardiac arrest caused by accidental severe hypothermia and myocardial infarction during general anesthesia.
Therapeutic hypothermia is often used for traumatic brain injury because of its neuroprotective effect and decreased secondary brain injury. However, this procedure lacks clinical evidence supporting its efficacy, and adverse outcomes have been reported during general anesthesia. A 61-year-old man with a history of percutaneous coronary intervention (PCI) was admitted with traumatic brain injury. Immediately after admission, he underwent mild therapeutic hypothermia with a target temperature of 33.0°C for neuroprotection. During general anesthesia for emergency surgery because he developed a mass effect, hypothermic cardiac arrest occurred following an additional decrease in the core body temperature. Moreover, myocardial infarction caused by restenosis of the previous PCI lesion also contributed to the cardiac arrest. Although the patient recovered spontaneous circulation after an hour-long cardiopulmonary resuscitation with rewarming, he eventually died of subsequent repetitive cardiac arrests. When anesthetizing patients undergoing therapeutic hypothermia, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.
Introduction
Therapeutic hypothermia is an attractive strategy for traumatic brain injury (TBI), which demonstrates neuroprotective effects and decreases secondary brain injury.1 Despite the lack of clinical evidence regarding efficacy, therapeutic hypothermia after TBI has been used for at least half a century because of its potential benefits, which are supported by numerous preclinical studies.2,3 However, there have been no reports on the danger of unintentional severe hypothermia during general anesthesia caused by the additional drop in temperature of patients undergoing therapeutic hypothermia. When anesthetizing patients with comorbidities, such as ischemic heart disease, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.4,5
Case report
A 61-year-old man (height, 168 cm; weight, 70 kg) presented to our emergency room (ER) with head trauma after falling on a cement floor from a 5-meter-tall construction site. Upon arrival at the ER, the patient had a Glasgow coma scale score of 11 and the following vital signs: blood pressure (BP), 147/83 mmHg; heart rate (HR), 86 beats per minute (bpm); respiratory rate, 24 breaths per minute; body temperature from the auditory canal, 36.8°C; and percutaneous oxygen saturation (SpO2), 97%. Upon discovering signs of acute subdural hemorrhage, subarachnoid hemorrhage, and intraventricular hemorrhage from a brain computed tomography (CT) scan performed in the ER, we admitted the patient to the intensive care unit (ICU) for close monitoring. After approximately 6 hours, we performed follow-up brain CT because the patient’s mental status had deteriorated from drowsy to stupor. Upon observing a mass effect characterized by an increase in the hemorrhage volume and a left midline shift on the CT images, the patient was scheduled for immediate intubation and emergency decompressive craniectomy (Figure 1).
Figure 1. Brain computed tomography (CT) images taken 6 hours after intensive care unit admission. (a) Axial view showing acute subarachnoid hemorrhage in the peri-mesencephalic cistern (arrow A) and subdural hemorrhage in the right inferior temporal (arrow B) and right parietal (arrow C) regions. (b) Axial view showing acute subdural hemorrhage in the right temporal region (black arrow) and left midline shift.
Although the patient had a history of ischemic heart disease with percutaneous coronary intervention (PCI), it was difficult to obtain accurate information because there was no relevant data in the patient’s medical records from the authors’ hospital, and his drowsy state interfered with detailed interviews. According to the chest CT scan performed immediately after admission, coronary stents had been inserted at the bifurcation site of the left main coronary artery (Figure 2). It also appeared that the inserted stents were drug-eluting stents (DES) considering that he had received dual antiplatelet therapy (DAPT) with aspirin and clopidogrel (Plavix®; Sanofi, Bridgewater, NJ, USA) a switching platelet P2Y12 inhibitor, for approximately 4 months. However, DAPT was discontinued owing to the worsening cerebral hemorrhage after hospitalization; bridging antiplatelet therapy was also not performed. Before the surgery, the patient showed a normal electrocardiogram (ECG), and cardiac biomarker testing revealed a troponin-T concentration of 0.006 µg/L, which was within normal limits (Figure 3). The patient demonstrated satisfactory functional capacity to be able to work at a construction site just before the trauma. Therefore, an additional preoperative cardiac evaluation was not considered necessary.
Figure 2. Chest computed tomography (CT) images showing coronary stents in the main bifurcation of the left coronary arteries (white arrow in (a) and (b)).
Figure 3. Preoperative electrocardiogram of the patient showing normal sinus rhythm with a heart rate of 95 bpm.
bpm, beats per minute.
Immediately after admission to the ICU, the patient’s bladder temperature was 38.0°C. Using a hydrogel-coated water-circulating pad (Arctic Sun®; Medivance Inc., Louisville, CO, USA), which is a surface cooling device, the patient underwent mild therapeutic hypothermia with a target temperature of 33.0°C. During therapeutic hypothermia, sedation and shivering control were performed using continuous rate infusions of midazolam at 5 mg/hour and remifentanil at 0.2 µg/kg/minute. The patient’s bladder temperature was 33.4°C just before he was transferred to the operating room. Preoperative laboratory tests showed no unusual findings except for mildly elevated aspartate aminotransferase (73 U/L) and alanine aminotransferase (51 U/L) concentrations.
The patient’s vital signs upon arrival at the operating room were a BP of 130/70 mmHg and SpO2 of 100%. Lead II of the ECG was monitored, and there were no unusual findings except sinus tachycardia with an HR of 100 bpm on the initial ECG. Arterial BP (ABP), central venous pressure (CVP), and esophageal temperature were monitored from the right radial artery, right subclavian vein, and lower third of the esophagus using a temperature sensor attached to the esophageal stethoscope, respectively. The initial ABP, CVP, and esophageal temperature were 140/85 mmHg, 6 mmHg, and 33.0°C, respectively. Arterial blood gas analysis was performed in the arterial line, with values calculated as for a body temperature of 37°C, which revealed the following: pH, 7.400; partial pressure of carbon dioxide (PCO2), 40 mmHg; partial pressure of oxygen (PO2), 174 mmHg; bicarbonate (HCO3), 24.8 mmol/L; base excess, 0.0; lactate, 0.17 mmol/L; ionized calcium, 1.04 mmol/L; and potassium, 3.3 mmol/L. Total intravenous (IV) anesthesia was performed using a target-controlled infusion of propofol and remifentanil in accordance with the Schnider6 and Minto7 models, respectively. Propofol and remifentanil were infused continuously with effect-site concentrations (Ce) of 1.5 µg/mL and 1.5 ng/mL, respectively. Concurrently, 50 mg of rocuronium was administered by IV bolus injection for neuromuscular blockade. Mechanical ventilation was adjusted to maintain the end-tidal carbon dioxide value at 30 to 33 mmHg.
As ABP decreased to 78/42 mmHg and CVP decreased to 4 mmHg within 10 minutes after anesthesia, the Ce of propofol and remifentanil were lowered to 0.8 µg/mL and 1.0 ng/mL, respectively, and approximately 1 L of balanced crystalloid fluid loading was performed. However, the patient’s BP did not recover, and a dopamine continuous infusion (5 µg/kg/minute) was initiated. When a surgical incision was made 10 minutes after beginning the dopamine infusion, the patient’s ABP and HR increased to 115/70 mmHg and 120 bpm, respectively, and marked ST-segment depression (−2.4 mm) was found, which was not observed on the preoperative or initial ECGs (Figure 4). Subsequently, the depth of anesthesia was increased by increasing the Ce of propofol to 1.0 to 1.2 µg/mL, followed by a continuous infusion of isosorbide dinitrate (1 µg/kg/minute). While HR was maintained at 110 to 120 bpm, ABP gradually decreased to 70/40 mmHg after approximately 5 minutes. To treat the hypotension, isosorbide dinitrate was discontinued, and a bolus injection of 100 µg phenylephrine was administered, after which, ABP increased temporarily to 95/60 mmHg. Subsequently, HR decreased suddenly and rapidly, and ventricular fibrillation occurred. The anesthesiologists immediately discontinued the surgery and performed cardiopulmonary resuscitation (CPR). The patient’s esophageal temperature at the time was 32.0°C.
Figure 4. Intraoperative electrocardiogram of the patient showing sinus tachycardia with significant ST depression. The ST depression on the monitor represented approximately −2.4 at the time.
Chest compressions were performed with periodic defibrillation with 200 J, and epinephrine injections were administered every 3 minutes; these measures elicited no response. During CPR, ECG monitoring showed temporary severe bradycardia (HR, 10–30 bpm). However, this was considered pulseless electrical activity because no pulse was detected in the patient’s femoral artery. Soon afterward, there was a rapid transition to ventricular fibrillation, and CPR was repeated. At the time, the patient’s esophageal temperature dropped to 30.0°C. According to the refractory state after the defibrillation and drug injections, we considered that the cardiac arrest was caused by hypothermia. To reverse the cause, a forced-air warming device and a warming blanket (Bair Hugger®; Arizant Healthcare Inc., Eden Prairie, MN, USA) were used for surface warming, the IV fluids were warmed, and surgical wound closure and dressing were performed rapidly to prevent surface heat loss. The first arterial blood gas analysis after ventricular fibrillation, calculated as for a body temperature of 37°C, showed the following: pH, 7.690; PCO2, 15 mmHg; PO2, 260 mmHg; HCO3, 18.1 mmol/L; base excess, −1.2; lactate, 0.33 mmol/L; ionized calcium, 0.88 mmol/L; and potassium, 5.0 mmol/L. Follow-up arterial blood gas analyses were repeated periodically. As we continued to perform CPR, the patient developed severe metabolic acidosis and hypocalcemia characterized by a pH of 6.840, HCO3 of 15.4 mmol/L, and ionized calcium of 0.66 mmol/L. To correct the abnormal findings, we administered appropriate amounts of sodium bicarbonate and calcium chloride. After an hour-long CPR and rewarming, the patient’s esophageal temperature recovered to 33.0°C, which was followed by observing sinus rhythm characterized by an HR of 75 bpm on an ECG monitor and detecting a pulse in the femoral artery with a characteristic ABP waveform that eventually led to the recovery of BP to 130/55 mmHg. Upon confirming recovery of spontaneous circulation (ROSC), the patient was transported to the ICU while we administered dopamine (10 µg/kg/minute), norepinephrine (0.1 µg/kg/minute), and vasopressin (2 U/hour). ABP, HR, and esophageal temperature during anesthesia are shown in Figure 5.
Figure 5. Changes in vital signs (blood pressure (BP), heart rate (HR), and esophageal temperature) over time (5-minute intervals).
bpm, beats per minute.
Cardiac biomarker testing performed after the patient’s transfer to the ICU showed an increased troponin-T level of 1.432 µg/L. Portable Doppler transthoracic echocardiography (TTE) revealed total occlusion of the left main coronary artery, which had undergone PCI previously. Because emergency coronary artery bypass grafting (CABG) was impossible at the time, repeated PCI was considered an alternative treatment option. However, even after his transfer to the ICU, the patient experienced repetitive ventricular fibrillation. Four hours after the patient’s transfer to the ICU, the patient’s guardian requested that CPR should be stopped. Without resuscitation, the patient died.
Discussion
In this case, the patient, who was receiving therapeutic hypothermia, underwent emergency craniectomy owing to worsening brain hemorrhage, and cardiac arrest occurred during general anesthesia. We deduced that the cardiac arrest was owing to the additional decrease in the patient’s core body temperature and myocardial infarction caused by restenosis of the previous PCI-treated lesion.
Therapeutic hypothermia is used to improve neurological outcomes in cardiac arrest, TBI, stroke, and hypoxic–ischemic encephalopathy.1,4 Studies on the efficacy of therapeutic hypothermia for TBI have been less convincing, and they show conflicting results. However, hypothermia is consistently used as an option for TBI because it reduces intracranial pressure and the cerebral metabolic rate, thereby protecting neuronal function in the brain and decreasing the wide range of injuries or reparative processes by attenuating temperature-sensitive mechanisms, such as excitotoxicity, free radical generation, apoptosis, and inflammation.1,2
The cardiovascular effects associated with hypothermia are complex. Hypothermia affects the electrical conduction of myocardial cells, which can induce hypotension or bradycardia and even increase the risk of ventricular fibrillation by prolonging repolarization.4,8 However, mild hypothermia of 32.0°C to 35.0°C demonstrates cardioprotective effects by decreasing myocardial oxygen demand and increasing myocardial perfusion through vasodilation of coronary vessels.9 Therefore, there has been a growing interest in the clinical use of mild therapeutic hypothermia as an adjunctive therapy during ischemic heart disease.10 Our patient received mild therapeutic hypothermia with a target core temperature of 33.0°C, which may have yielded cardioprotective as well as neuroprotective benefits.11
During general anesthesia, hypothermia is induced following an increase in heat loss caused by peripheral vasodilation after the injection of anesthetic drugs, heat redistribution from the core to the periphery, and impaired normal thermoregulatory responses, such as vasoconstriction or shivering.5 In our patient, the cold operating room temperature and irrigating an uncovered surgical area in the brain with cold saline also may have contributed to the additional core body temperature drop in addition to the effects of general anesthesia.12 However, in our case, neurosurgeons requested that anesthesiologists not perform warming during anesthesia. Instead, esophageal temperature and signs that could result from hypothermia, such as ECG changes, were carefully monitored. The esophageal temperature decreased from 33.0°C, which was measured immediately after entering the operation room, to 32.0°C at the time of cardiac arrest. The esophageal temperature further decreased to a minimum of 30.0°C during CPR. With rewarming, the temperature recovered to 33.0°C and finally, ROSC was achieved. Considering that chest compressions, drugs, and defibrillation did not induce ROSC, which was achieved with the recovery of an esophageal temperature of 33.0°C, the additional rapid fall in the patient’s core body temperature during anesthesia may have caused the cardiac arrest.
In this case study, we suspected that restenosis at the site of the previous stent insertion contributed to the cardiac arrest, given the following three findings: First, the patient had a normal preoperative ECG with no signs of ischemia but developed ST-segment depression during anesthesia. Second, repetitive cardiac arrests occurred even after achieving ROSC with rewarming. Third, an increased troponin-T concentration was observed, and suspected re-stenotic occlusion of the left main coronary artery was confirmed by Doppler TTE after the patient’s transfer to the ICU. Accordingly, myocardial infarction was considered the main cause of death.
The mechanisms of myocardial infarction owing to stent restenosis after DES insertion are rapid stent thrombosis or profuse in-stent restenosis (ISR).13 We assume that the lesion characteristics in the left main bifurcation of the coronary artery and DAPT interruption owing to the increasing amount of cerebral hemorrhage contributed to the stent restenosis in this patient. As evidence in support of this assumption, previous studies have shown that stenting at the left main bifurcation and DAPT interruption are strong risk factors for stent thrombosis or ISR.14,15 While controversial, not interrupting aspirin perioperatively could have avoided poor outcomes.16 However, there were no clinical signs related to expected ongoing restenosis of the coronary artery or myocardial ischemia preoperatively in our patient. We suspect that the possible ischemic signs may have been masked by the aforementioned cardioprotective benefits of therapeutic hypothermia. We also believe that the following three conditions contributed to the intraoperative myocardial ischemia and infarction caused by acute restenosis of the coronary stent: First, intraoperative hypotension may have contributed to myocardial injury. The IV anesthetics used during anesthesia, propofol and remifentanil, induced redistribution of blood volume by peripheral vasodilation, resulting in profound hypotension.4 Profound hypotension was strongly associated with myocardial injury and especially harmful for this patient, who had preexisting ischemic heart disease.17 Second, tachycardia, which may have been caused by the inotrope dopamine, used to compensate for the low BP or the stress response to surgical stimulation from shallow anesthetic depth, may have triggered myocardial ischemia by decreasing oxygen delivery and increasing oxygen consumption.17,18 If cardiogenic shock is suspected, as in our case, norepinephrine could be a better option than dopamine.19 Third, the bolus injection of phenylephrine, a selective α1-adrenergic agonist, just before cardiac arrest may have temporarily and abruptly increased cardiac afterload and impaired myocardial perfusion.20
Had we known that coronary artery occlusion would result from hypothermia-induced cardiac arrest and coronary stent restenosis, the treatment of choice would have been CABG surgery while maintaining circulation with emergency cardiopulmonary bypass (CPB) or extracorporeal membrane oxygenation (ECMO) via the femoral artery and vein.11 However, it was difficult to determine the etiology of the cardiac arrest at the time. Even if we had known the cause, ideal treatment would have been impossible, given circumstances beyond our control; all thoracic surgeons were busy operating, and emergency CPB and ECMO were not available. As an alternative, repeat PCI after the patient achieved ROSC could have been performed. However, the patient had a prolonged CPR duration, and he did not maintain the ROSC state for a period sufficient for intervention, which led to a poor outcome.
In this case, it was not possible to accurately determine the reason for the cardiac arrest and for not achieving consistent ROSC. Instead, we conclude that cardiac arrest developed and persisted as a result of the unintentional severe hypothermia that occurred during general anesthesia and/or coronary artery restenosis at the site of the previous PCI. A thorough evaluation of heart-related history is required for TBI patients who have a history of ischemic heart diseases and who undergo therapeutic hypothermia because there may be undiscovered coronary insufficiency owing to the cardioprotective effects of therapeutic hypothermia. In addition, careful monitoring of the patient’s core body temperature, and preventive and countermeasures should be fully prepared and taken against adverse hemodynamic effects, even cardiac arrest, which can occur because of additive hypothermic responses during anesthesia.
Ethics statement: This case report was approved by our Institutional Review Board (EMC 2020-05-005-001), and the case report was performed in compliance with the EQUATOR Network guidelines (the CARE guidelines). All patient details were de-identified, and written consent was obtained from the patient’s guardian for treatment and publication of this report.
Declaration of conflicting interest: The authors declare that there is no conflict of interest.
Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
ORCID iDs: Dong Ho Park https://orcid.org/0000-0002-6587-3756
Tae Woo Kim https://orcid.org/0000-0002-2115-6448
Mo Se Kim https://orcid.org/0000-0001-7362-0030 | Recovered | ReactionOutcome | CC BY-NC | 33499678 | 19,429,456 | 2021-01 |
What was the outcome of reaction 'Hypothermia'? | Cardiac arrest caused by accidental severe hypothermia and myocardial infarction during general anesthesia.
Therapeutic hypothermia is often used for traumatic brain injury because of its neuroprotective effect and decreased secondary brain injury. However, this procedure lacks clinical evidence supporting its efficacy, and adverse outcomes have been reported during general anesthesia. A 61-year-old man with a history of percutaneous coronary intervention (PCI) was admitted with traumatic brain injury. Immediately after admission, he underwent mild therapeutic hypothermia with a target temperature of 33.0°C for neuroprotection. During general anesthesia for emergency surgery because he developed a mass effect, hypothermic cardiac arrest occurred following an additional decrease in the core body temperature. Moreover, myocardial infarction caused by restenosis of the previous PCI lesion also contributed to the cardiac arrest. Although the patient recovered spontaneous circulation after an hour-long cardiopulmonary resuscitation with rewarming, he eventually died of subsequent repetitive cardiac arrests. When anesthetizing patients undergoing therapeutic hypothermia, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.
Introduction
Therapeutic hypothermia is an attractive strategy for traumatic brain injury (TBI), which demonstrates neuroprotective effects and decreases secondary brain injury.1 Despite the lack of clinical evidence regarding efficacy, therapeutic hypothermia after TBI has been used for at least half a century because of its potential benefits, which are supported by numerous preclinical studies.2,3 However, there have been no reports on the danger of unintentional severe hypothermia during general anesthesia caused by the additional drop in temperature of patients undergoing therapeutic hypothermia. When anesthetizing patients with comorbidities, such as ischemic heart disease, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.4,5
Case report
A 61-year-old man (height, 168 cm; weight, 70 kg) presented to our emergency room (ER) with head trauma after falling on a cement floor from a 5-meter-tall construction site. Upon arrival at the ER, the patient had a Glasgow coma scale score of 11 and the following vital signs: blood pressure (BP), 147/83 mmHg; heart rate (HR), 86 beats per minute (bpm); respiratory rate, 24 breaths per minute; body temperature from the auditory canal, 36.8°C; and percutaneous oxygen saturation (SpO2), 97%. Upon discovering signs of acute subdural hemorrhage, subarachnoid hemorrhage, and intraventricular hemorrhage from a brain computed tomography (CT) scan performed in the ER, we admitted the patient to the intensive care unit (ICU) for close monitoring. After approximately 6 hours, we performed follow-up brain CT because the patient’s mental status had deteriorated from drowsy to stupor. Upon observing a mass effect characterized by an increase in the hemorrhage volume and a left midline shift on the CT images, the patient was scheduled for immediate intubation and emergency decompressive craniectomy (Figure 1).
Figure 1. Brain computed tomography (CT) images taken 6 hours after intensive care unit admission. (a) Axial view showing acute subarachnoid hemorrhage in the peri-mesencephalic cistern (arrow A) and subdural hemorrhage in the right inferior temporal (arrow B) and right parietal (arrow C) regions. (b) Axial view showing acute subdural hemorrhage in the right temporal region (black arrow) and left midline shift.
Although the patient had a history of ischemic heart disease with percutaneous coronary intervention (PCI), it was difficult to obtain accurate information because there was no relevant data in the patient’s medical records from the authors’ hospital, and his drowsy state interfered with detailed interviews. According to the chest CT scan performed immediately after admission, coronary stents had been inserted at the bifurcation site of the left main coronary artery (Figure 2). It also appeared that the inserted stents were drug-eluting stents (DES) considering that he had received dual antiplatelet therapy (DAPT) with aspirin and clopidogrel (Plavix®; Sanofi, Bridgewater, NJ, USA) a switching platelet P2Y12 inhibitor, for approximately 4 months. However, DAPT was discontinued owing to the worsening cerebral hemorrhage after hospitalization; bridging antiplatelet therapy was also not performed. Before the surgery, the patient showed a normal electrocardiogram (ECG), and cardiac biomarker testing revealed a troponin-T concentration of 0.006 µg/L, which was within normal limits (Figure 3). The patient demonstrated satisfactory functional capacity to be able to work at a construction site just before the trauma. Therefore, an additional preoperative cardiac evaluation was not considered necessary.
Figure 2. Chest computed tomography (CT) images showing coronary stents in the main bifurcation of the left coronary arteries (white arrow in (a) and (b)).
Figure 3. Preoperative electrocardiogram of the patient showing normal sinus rhythm with a heart rate of 95 bpm.
bpm, beats per minute.
Immediately after admission to the ICU, the patient’s bladder temperature was 38.0°C. Using a hydrogel-coated water-circulating pad (Arctic Sun®; Medivance Inc., Louisville, CO, USA), which is a surface cooling device, the patient underwent mild therapeutic hypothermia with a target temperature of 33.0°C. During therapeutic hypothermia, sedation and shivering control were performed using continuous rate infusions of midazolam at 5 mg/hour and remifentanil at 0.2 µg/kg/minute. The patient’s bladder temperature was 33.4°C just before he was transferred to the operating room. Preoperative laboratory tests showed no unusual findings except for mildly elevated aspartate aminotransferase (73 U/L) and alanine aminotransferase (51 U/L) concentrations.
The patient’s vital signs upon arrival at the operating room were a BP of 130/70 mmHg and SpO2 of 100%. Lead II of the ECG was monitored, and there were no unusual findings except sinus tachycardia with an HR of 100 bpm on the initial ECG. Arterial BP (ABP), central venous pressure (CVP), and esophageal temperature were monitored from the right radial artery, right subclavian vein, and lower third of the esophagus using a temperature sensor attached to the esophageal stethoscope, respectively. The initial ABP, CVP, and esophageal temperature were 140/85 mmHg, 6 mmHg, and 33.0°C, respectively. Arterial blood gas analysis was performed in the arterial line, with values calculated as for a body temperature of 37°C, which revealed the following: pH, 7.400; partial pressure of carbon dioxide (PCO2), 40 mmHg; partial pressure of oxygen (PO2), 174 mmHg; bicarbonate (HCO3), 24.8 mmol/L; base excess, 0.0; lactate, 0.17 mmol/L; ionized calcium, 1.04 mmol/L; and potassium, 3.3 mmol/L. Total intravenous (IV) anesthesia was performed using a target-controlled infusion of propofol and remifentanil in accordance with the Schnider6 and Minto7 models, respectively. Propofol and remifentanil were infused continuously with effect-site concentrations (Ce) of 1.5 µg/mL and 1.5 ng/mL, respectively. Concurrently, 50 mg of rocuronium was administered by IV bolus injection for neuromuscular blockade. Mechanical ventilation was adjusted to maintain the end-tidal carbon dioxide value at 30 to 33 mmHg.
As ABP decreased to 78/42 mmHg and CVP decreased to 4 mmHg within 10 minutes after anesthesia, the Ce of propofol and remifentanil were lowered to 0.8 µg/mL and 1.0 ng/mL, respectively, and approximately 1 L of balanced crystalloid fluid loading was performed. However, the patient’s BP did not recover, and a dopamine continuous infusion (5 µg/kg/minute) was initiated. When a surgical incision was made 10 minutes after beginning the dopamine infusion, the patient’s ABP and HR increased to 115/70 mmHg and 120 bpm, respectively, and marked ST-segment depression (−2.4 mm) was found, which was not observed on the preoperative or initial ECGs (Figure 4). Subsequently, the depth of anesthesia was increased by increasing the Ce of propofol to 1.0 to 1.2 µg/mL, followed by a continuous infusion of isosorbide dinitrate (1 µg/kg/minute). While HR was maintained at 110 to 120 bpm, ABP gradually decreased to 70/40 mmHg after approximately 5 minutes. To treat the hypotension, isosorbide dinitrate was discontinued, and a bolus injection of 100 µg phenylephrine was administered, after which, ABP increased temporarily to 95/60 mmHg. Subsequently, HR decreased suddenly and rapidly, and ventricular fibrillation occurred. The anesthesiologists immediately discontinued the surgery and performed cardiopulmonary resuscitation (CPR). The patient’s esophageal temperature at the time was 32.0°C.
Figure 4. Intraoperative electrocardiogram of the patient showing sinus tachycardia with significant ST depression. The ST depression on the monitor represented approximately −2.4 at the time.
Chest compressions were performed with periodic defibrillation with 200 J, and epinephrine injections were administered every 3 minutes; these measures elicited no response. During CPR, ECG monitoring showed temporary severe bradycardia (HR, 10–30 bpm). However, this was considered pulseless electrical activity because no pulse was detected in the patient’s femoral artery. Soon afterward, there was a rapid transition to ventricular fibrillation, and CPR was repeated. At the time, the patient’s esophageal temperature dropped to 30.0°C. According to the refractory state after the defibrillation and drug injections, we considered that the cardiac arrest was caused by hypothermia. To reverse the cause, a forced-air warming device and a warming blanket (Bair Hugger®; Arizant Healthcare Inc., Eden Prairie, MN, USA) were used for surface warming, the IV fluids were warmed, and surgical wound closure and dressing were performed rapidly to prevent surface heat loss. The first arterial blood gas analysis after ventricular fibrillation, calculated as for a body temperature of 37°C, showed the following: pH, 7.690; PCO2, 15 mmHg; PO2, 260 mmHg; HCO3, 18.1 mmol/L; base excess, −1.2; lactate, 0.33 mmol/L; ionized calcium, 0.88 mmol/L; and potassium, 5.0 mmol/L. Follow-up arterial blood gas analyses were repeated periodically. As we continued to perform CPR, the patient developed severe metabolic acidosis and hypocalcemia characterized by a pH of 6.840, HCO3 of 15.4 mmol/L, and ionized calcium of 0.66 mmol/L. To correct the abnormal findings, we administered appropriate amounts of sodium bicarbonate and calcium chloride. After an hour-long CPR and rewarming, the patient’s esophageal temperature recovered to 33.0°C, which was followed by observing sinus rhythm characterized by an HR of 75 bpm on an ECG monitor and detecting a pulse in the femoral artery with a characteristic ABP waveform that eventually led to the recovery of BP to 130/55 mmHg. Upon confirming recovery of spontaneous circulation (ROSC), the patient was transported to the ICU while we administered dopamine (10 µg/kg/minute), norepinephrine (0.1 µg/kg/minute), and vasopressin (2 U/hour). ABP, HR, and esophageal temperature during anesthesia are shown in Figure 5.
Figure 5. Changes in vital signs (blood pressure (BP), heart rate (HR), and esophageal temperature) over time (5-minute intervals).
bpm, beats per minute.
Cardiac biomarker testing performed after the patient’s transfer to the ICU showed an increased troponin-T level of 1.432 µg/L. Portable Doppler transthoracic echocardiography (TTE) revealed total occlusion of the left main coronary artery, which had undergone PCI previously. Because emergency coronary artery bypass grafting (CABG) was impossible at the time, repeated PCI was considered an alternative treatment option. However, even after his transfer to the ICU, the patient experienced repetitive ventricular fibrillation. Four hours after the patient’s transfer to the ICU, the patient’s guardian requested that CPR should be stopped. Without resuscitation, the patient died.
Discussion
In this case, the patient, who was receiving therapeutic hypothermia, underwent emergency craniectomy owing to worsening brain hemorrhage, and cardiac arrest occurred during general anesthesia. We deduced that the cardiac arrest was owing to the additional decrease in the patient’s core body temperature and myocardial infarction caused by restenosis of the previous PCI-treated lesion.
Therapeutic hypothermia is used to improve neurological outcomes in cardiac arrest, TBI, stroke, and hypoxic–ischemic encephalopathy.1,4 Studies on the efficacy of therapeutic hypothermia for TBI have been less convincing, and they show conflicting results. However, hypothermia is consistently used as an option for TBI because it reduces intracranial pressure and the cerebral metabolic rate, thereby protecting neuronal function in the brain and decreasing the wide range of injuries or reparative processes by attenuating temperature-sensitive mechanisms, such as excitotoxicity, free radical generation, apoptosis, and inflammation.1,2
The cardiovascular effects associated with hypothermia are complex. Hypothermia affects the electrical conduction of myocardial cells, which can induce hypotension or bradycardia and even increase the risk of ventricular fibrillation by prolonging repolarization.4,8 However, mild hypothermia of 32.0°C to 35.0°C demonstrates cardioprotective effects by decreasing myocardial oxygen demand and increasing myocardial perfusion through vasodilation of coronary vessels.9 Therefore, there has been a growing interest in the clinical use of mild therapeutic hypothermia as an adjunctive therapy during ischemic heart disease.10 Our patient received mild therapeutic hypothermia with a target core temperature of 33.0°C, which may have yielded cardioprotective as well as neuroprotective benefits.11
During general anesthesia, hypothermia is induced following an increase in heat loss caused by peripheral vasodilation after the injection of anesthetic drugs, heat redistribution from the core to the periphery, and impaired normal thermoregulatory responses, such as vasoconstriction or shivering.5 In our patient, the cold operating room temperature and irrigating an uncovered surgical area in the brain with cold saline also may have contributed to the additional core body temperature drop in addition to the effects of general anesthesia.12 However, in our case, neurosurgeons requested that anesthesiologists not perform warming during anesthesia. Instead, esophageal temperature and signs that could result from hypothermia, such as ECG changes, were carefully monitored. The esophageal temperature decreased from 33.0°C, which was measured immediately after entering the operation room, to 32.0°C at the time of cardiac arrest. The esophageal temperature further decreased to a minimum of 30.0°C during CPR. With rewarming, the temperature recovered to 33.0°C and finally, ROSC was achieved. Considering that chest compressions, drugs, and defibrillation did not induce ROSC, which was achieved with the recovery of an esophageal temperature of 33.0°C, the additional rapid fall in the patient’s core body temperature during anesthesia may have caused the cardiac arrest.
In this case study, we suspected that restenosis at the site of the previous stent insertion contributed to the cardiac arrest, given the following three findings: First, the patient had a normal preoperative ECG with no signs of ischemia but developed ST-segment depression during anesthesia. Second, repetitive cardiac arrests occurred even after achieving ROSC with rewarming. Third, an increased troponin-T concentration was observed, and suspected re-stenotic occlusion of the left main coronary artery was confirmed by Doppler TTE after the patient’s transfer to the ICU. Accordingly, myocardial infarction was considered the main cause of death.
The mechanisms of myocardial infarction owing to stent restenosis after DES insertion are rapid stent thrombosis or profuse in-stent restenosis (ISR).13 We assume that the lesion characteristics in the left main bifurcation of the coronary artery and DAPT interruption owing to the increasing amount of cerebral hemorrhage contributed to the stent restenosis in this patient. As evidence in support of this assumption, previous studies have shown that stenting at the left main bifurcation and DAPT interruption are strong risk factors for stent thrombosis or ISR.14,15 While controversial, not interrupting aspirin perioperatively could have avoided poor outcomes.16 However, there were no clinical signs related to expected ongoing restenosis of the coronary artery or myocardial ischemia preoperatively in our patient. We suspect that the possible ischemic signs may have been masked by the aforementioned cardioprotective benefits of therapeutic hypothermia. We also believe that the following three conditions contributed to the intraoperative myocardial ischemia and infarction caused by acute restenosis of the coronary stent: First, intraoperative hypotension may have contributed to myocardial injury. The IV anesthetics used during anesthesia, propofol and remifentanil, induced redistribution of blood volume by peripheral vasodilation, resulting in profound hypotension.4 Profound hypotension was strongly associated with myocardial injury and especially harmful for this patient, who had preexisting ischemic heart disease.17 Second, tachycardia, which may have been caused by the inotrope dopamine, used to compensate for the low BP or the stress response to surgical stimulation from shallow anesthetic depth, may have triggered myocardial ischemia by decreasing oxygen delivery and increasing oxygen consumption.17,18 If cardiogenic shock is suspected, as in our case, norepinephrine could be a better option than dopamine.19 Third, the bolus injection of phenylephrine, a selective α1-adrenergic agonist, just before cardiac arrest may have temporarily and abruptly increased cardiac afterload and impaired myocardial perfusion.20
Had we known that coronary artery occlusion would result from hypothermia-induced cardiac arrest and coronary stent restenosis, the treatment of choice would have been CABG surgery while maintaining circulation with emergency cardiopulmonary bypass (CPB) or extracorporeal membrane oxygenation (ECMO) via the femoral artery and vein.11 However, it was difficult to determine the etiology of the cardiac arrest at the time. Even if we had known the cause, ideal treatment would have been impossible, given circumstances beyond our control; all thoracic surgeons were busy operating, and emergency CPB and ECMO were not available. As an alternative, repeat PCI after the patient achieved ROSC could have been performed. However, the patient had a prolonged CPR duration, and he did not maintain the ROSC state for a period sufficient for intervention, which led to a poor outcome.
In this case, it was not possible to accurately determine the reason for the cardiac arrest and for not achieving consistent ROSC. Instead, we conclude that cardiac arrest developed and persisted as a result of the unintentional severe hypothermia that occurred during general anesthesia and/or coronary artery restenosis at the site of the previous PCI. A thorough evaluation of heart-related history is required for TBI patients who have a history of ischemic heart diseases and who undergo therapeutic hypothermia because there may be undiscovered coronary insufficiency owing to the cardioprotective effects of therapeutic hypothermia. In addition, careful monitoring of the patient’s core body temperature, and preventive and countermeasures should be fully prepared and taken against adverse hemodynamic effects, even cardiac arrest, which can occur because of additive hypothermic responses during anesthesia.
Ethics statement: This case report was approved by our Institutional Review Board (EMC 2020-05-005-001), and the case report was performed in compliance with the EQUATOR Network guidelines (the CARE guidelines). All patient details were de-identified, and written consent was obtained from the patient’s guardian for treatment and publication of this report.
Declaration of conflicting interest: The authors declare that there is no conflict of interest.
Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
ORCID iDs: Dong Ho Park https://orcid.org/0000-0002-6587-3756
Tae Woo Kim https://orcid.org/0000-0002-2115-6448
Mo Se Kim https://orcid.org/0000-0001-7362-0030 | Recovered | ReactionOutcome | CC BY-NC | 33499678 | 19,429,456 | 2021-01 |
What was the outcome of reaction 'Ischaemia'? | Cardiac arrest caused by accidental severe hypothermia and myocardial infarction during general anesthesia.
Therapeutic hypothermia is often used for traumatic brain injury because of its neuroprotective effect and decreased secondary brain injury. However, this procedure lacks clinical evidence supporting its efficacy, and adverse outcomes have been reported during general anesthesia. A 61-year-old man with a history of percutaneous coronary intervention (PCI) was admitted with traumatic brain injury. Immediately after admission, he underwent mild therapeutic hypothermia with a target temperature of 33.0°C for neuroprotection. During general anesthesia for emergency surgery because he developed a mass effect, hypothermic cardiac arrest occurred following an additional decrease in the core body temperature. Moreover, myocardial infarction caused by restenosis of the previous PCI lesion also contributed to the cardiac arrest. Although the patient recovered spontaneous circulation after an hour-long cardiopulmonary resuscitation with rewarming, he eventually died of subsequent repetitive cardiac arrests. When anesthetizing patients undergoing therapeutic hypothermia, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.
Introduction
Therapeutic hypothermia is an attractive strategy for traumatic brain injury (TBI), which demonstrates neuroprotective effects and decreases secondary brain injury.1 Despite the lack of clinical evidence regarding efficacy, therapeutic hypothermia after TBI has been used for at least half a century because of its potential benefits, which are supported by numerous preclinical studies.2,3 However, there have been no reports on the danger of unintentional severe hypothermia during general anesthesia caused by the additional drop in temperature of patients undergoing therapeutic hypothermia. When anesthetizing patients with comorbidities, such as ischemic heart disease, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.4,5
Case report
A 61-year-old man (height, 168 cm; weight, 70 kg) presented to our emergency room (ER) with head trauma after falling on a cement floor from a 5-meter-tall construction site. Upon arrival at the ER, the patient had a Glasgow coma scale score of 11 and the following vital signs: blood pressure (BP), 147/83 mmHg; heart rate (HR), 86 beats per minute (bpm); respiratory rate, 24 breaths per minute; body temperature from the auditory canal, 36.8°C; and percutaneous oxygen saturation (SpO2), 97%. Upon discovering signs of acute subdural hemorrhage, subarachnoid hemorrhage, and intraventricular hemorrhage from a brain computed tomography (CT) scan performed in the ER, we admitted the patient to the intensive care unit (ICU) for close monitoring. After approximately 6 hours, we performed follow-up brain CT because the patient’s mental status had deteriorated from drowsy to stupor. Upon observing a mass effect characterized by an increase in the hemorrhage volume and a left midline shift on the CT images, the patient was scheduled for immediate intubation and emergency decompressive craniectomy (Figure 1).
Figure 1. Brain computed tomography (CT) images taken 6 hours after intensive care unit admission. (a) Axial view showing acute subarachnoid hemorrhage in the peri-mesencephalic cistern (arrow A) and subdural hemorrhage in the right inferior temporal (arrow B) and right parietal (arrow C) regions. (b) Axial view showing acute subdural hemorrhage in the right temporal region (black arrow) and left midline shift.
Although the patient had a history of ischemic heart disease with percutaneous coronary intervention (PCI), it was difficult to obtain accurate information because there was no relevant data in the patient’s medical records from the authors’ hospital, and his drowsy state interfered with detailed interviews. According to the chest CT scan performed immediately after admission, coronary stents had been inserted at the bifurcation site of the left main coronary artery (Figure 2). It also appeared that the inserted stents were drug-eluting stents (DES) considering that he had received dual antiplatelet therapy (DAPT) with aspirin and clopidogrel (Plavix®; Sanofi, Bridgewater, NJ, USA) a switching platelet P2Y12 inhibitor, for approximately 4 months. However, DAPT was discontinued owing to the worsening cerebral hemorrhage after hospitalization; bridging antiplatelet therapy was also not performed. Before the surgery, the patient showed a normal electrocardiogram (ECG), and cardiac biomarker testing revealed a troponin-T concentration of 0.006 µg/L, which was within normal limits (Figure 3). The patient demonstrated satisfactory functional capacity to be able to work at a construction site just before the trauma. Therefore, an additional preoperative cardiac evaluation was not considered necessary.
Figure 2. Chest computed tomography (CT) images showing coronary stents in the main bifurcation of the left coronary arteries (white arrow in (a) and (b)).
Figure 3. Preoperative electrocardiogram of the patient showing normal sinus rhythm with a heart rate of 95 bpm.
bpm, beats per minute.
Immediately after admission to the ICU, the patient’s bladder temperature was 38.0°C. Using a hydrogel-coated water-circulating pad (Arctic Sun®; Medivance Inc., Louisville, CO, USA), which is a surface cooling device, the patient underwent mild therapeutic hypothermia with a target temperature of 33.0°C. During therapeutic hypothermia, sedation and shivering control were performed using continuous rate infusions of midazolam at 5 mg/hour and remifentanil at 0.2 µg/kg/minute. The patient’s bladder temperature was 33.4°C just before he was transferred to the operating room. Preoperative laboratory tests showed no unusual findings except for mildly elevated aspartate aminotransferase (73 U/L) and alanine aminotransferase (51 U/L) concentrations.
The patient’s vital signs upon arrival at the operating room were a BP of 130/70 mmHg and SpO2 of 100%. Lead II of the ECG was monitored, and there were no unusual findings except sinus tachycardia with an HR of 100 bpm on the initial ECG. Arterial BP (ABP), central venous pressure (CVP), and esophageal temperature were monitored from the right radial artery, right subclavian vein, and lower third of the esophagus using a temperature sensor attached to the esophageal stethoscope, respectively. The initial ABP, CVP, and esophageal temperature were 140/85 mmHg, 6 mmHg, and 33.0°C, respectively. Arterial blood gas analysis was performed in the arterial line, with values calculated as for a body temperature of 37°C, which revealed the following: pH, 7.400; partial pressure of carbon dioxide (PCO2), 40 mmHg; partial pressure of oxygen (PO2), 174 mmHg; bicarbonate (HCO3), 24.8 mmol/L; base excess, 0.0; lactate, 0.17 mmol/L; ionized calcium, 1.04 mmol/L; and potassium, 3.3 mmol/L. Total intravenous (IV) anesthesia was performed using a target-controlled infusion of propofol and remifentanil in accordance with the Schnider6 and Minto7 models, respectively. Propofol and remifentanil were infused continuously with effect-site concentrations (Ce) of 1.5 µg/mL and 1.5 ng/mL, respectively. Concurrently, 50 mg of rocuronium was administered by IV bolus injection for neuromuscular blockade. Mechanical ventilation was adjusted to maintain the end-tidal carbon dioxide value at 30 to 33 mmHg.
As ABP decreased to 78/42 mmHg and CVP decreased to 4 mmHg within 10 minutes after anesthesia, the Ce of propofol and remifentanil were lowered to 0.8 µg/mL and 1.0 ng/mL, respectively, and approximately 1 L of balanced crystalloid fluid loading was performed. However, the patient’s BP did not recover, and a dopamine continuous infusion (5 µg/kg/minute) was initiated. When a surgical incision was made 10 minutes after beginning the dopamine infusion, the patient’s ABP and HR increased to 115/70 mmHg and 120 bpm, respectively, and marked ST-segment depression (−2.4 mm) was found, which was not observed on the preoperative or initial ECGs (Figure 4). Subsequently, the depth of anesthesia was increased by increasing the Ce of propofol to 1.0 to 1.2 µg/mL, followed by a continuous infusion of isosorbide dinitrate (1 µg/kg/minute). While HR was maintained at 110 to 120 bpm, ABP gradually decreased to 70/40 mmHg after approximately 5 minutes. To treat the hypotension, isosorbide dinitrate was discontinued, and a bolus injection of 100 µg phenylephrine was administered, after which, ABP increased temporarily to 95/60 mmHg. Subsequently, HR decreased suddenly and rapidly, and ventricular fibrillation occurred. The anesthesiologists immediately discontinued the surgery and performed cardiopulmonary resuscitation (CPR). The patient’s esophageal temperature at the time was 32.0°C.
Figure 4. Intraoperative electrocardiogram of the patient showing sinus tachycardia with significant ST depression. The ST depression on the monitor represented approximately −2.4 at the time.
Chest compressions were performed with periodic defibrillation with 200 J, and epinephrine injections were administered every 3 minutes; these measures elicited no response. During CPR, ECG monitoring showed temporary severe bradycardia (HR, 10–30 bpm). However, this was considered pulseless electrical activity because no pulse was detected in the patient’s femoral artery. Soon afterward, there was a rapid transition to ventricular fibrillation, and CPR was repeated. At the time, the patient’s esophageal temperature dropped to 30.0°C. According to the refractory state after the defibrillation and drug injections, we considered that the cardiac arrest was caused by hypothermia. To reverse the cause, a forced-air warming device and a warming blanket (Bair Hugger®; Arizant Healthcare Inc., Eden Prairie, MN, USA) were used for surface warming, the IV fluids were warmed, and surgical wound closure and dressing were performed rapidly to prevent surface heat loss. The first arterial blood gas analysis after ventricular fibrillation, calculated as for a body temperature of 37°C, showed the following: pH, 7.690; PCO2, 15 mmHg; PO2, 260 mmHg; HCO3, 18.1 mmol/L; base excess, −1.2; lactate, 0.33 mmol/L; ionized calcium, 0.88 mmol/L; and potassium, 5.0 mmol/L. Follow-up arterial blood gas analyses were repeated periodically. As we continued to perform CPR, the patient developed severe metabolic acidosis and hypocalcemia characterized by a pH of 6.840, HCO3 of 15.4 mmol/L, and ionized calcium of 0.66 mmol/L. To correct the abnormal findings, we administered appropriate amounts of sodium bicarbonate and calcium chloride. After an hour-long CPR and rewarming, the patient’s esophageal temperature recovered to 33.0°C, which was followed by observing sinus rhythm characterized by an HR of 75 bpm on an ECG monitor and detecting a pulse in the femoral artery with a characteristic ABP waveform that eventually led to the recovery of BP to 130/55 mmHg. Upon confirming recovery of spontaneous circulation (ROSC), the patient was transported to the ICU while we administered dopamine (10 µg/kg/minute), norepinephrine (0.1 µg/kg/minute), and vasopressin (2 U/hour). ABP, HR, and esophageal temperature during anesthesia are shown in Figure 5.
Figure 5. Changes in vital signs (blood pressure (BP), heart rate (HR), and esophageal temperature) over time (5-minute intervals).
bpm, beats per minute.
Cardiac biomarker testing performed after the patient’s transfer to the ICU showed an increased troponin-T level of 1.432 µg/L. Portable Doppler transthoracic echocardiography (TTE) revealed total occlusion of the left main coronary artery, which had undergone PCI previously. Because emergency coronary artery bypass grafting (CABG) was impossible at the time, repeated PCI was considered an alternative treatment option. However, even after his transfer to the ICU, the patient experienced repetitive ventricular fibrillation. Four hours after the patient’s transfer to the ICU, the patient’s guardian requested that CPR should be stopped. Without resuscitation, the patient died.
Discussion
In this case, the patient, who was receiving therapeutic hypothermia, underwent emergency craniectomy owing to worsening brain hemorrhage, and cardiac arrest occurred during general anesthesia. We deduced that the cardiac arrest was owing to the additional decrease in the patient’s core body temperature and myocardial infarction caused by restenosis of the previous PCI-treated lesion.
Therapeutic hypothermia is used to improve neurological outcomes in cardiac arrest, TBI, stroke, and hypoxic–ischemic encephalopathy.1,4 Studies on the efficacy of therapeutic hypothermia for TBI have been less convincing, and they show conflicting results. However, hypothermia is consistently used as an option for TBI because it reduces intracranial pressure and the cerebral metabolic rate, thereby protecting neuronal function in the brain and decreasing the wide range of injuries or reparative processes by attenuating temperature-sensitive mechanisms, such as excitotoxicity, free radical generation, apoptosis, and inflammation.1,2
The cardiovascular effects associated with hypothermia are complex. Hypothermia affects the electrical conduction of myocardial cells, which can induce hypotension or bradycardia and even increase the risk of ventricular fibrillation by prolonging repolarization.4,8 However, mild hypothermia of 32.0°C to 35.0°C demonstrates cardioprotective effects by decreasing myocardial oxygen demand and increasing myocardial perfusion through vasodilation of coronary vessels.9 Therefore, there has been a growing interest in the clinical use of mild therapeutic hypothermia as an adjunctive therapy during ischemic heart disease.10 Our patient received mild therapeutic hypothermia with a target core temperature of 33.0°C, which may have yielded cardioprotective as well as neuroprotective benefits.11
During general anesthesia, hypothermia is induced following an increase in heat loss caused by peripheral vasodilation after the injection of anesthetic drugs, heat redistribution from the core to the periphery, and impaired normal thermoregulatory responses, such as vasoconstriction or shivering.5 In our patient, the cold operating room temperature and irrigating an uncovered surgical area in the brain with cold saline also may have contributed to the additional core body temperature drop in addition to the effects of general anesthesia.12 However, in our case, neurosurgeons requested that anesthesiologists not perform warming during anesthesia. Instead, esophageal temperature and signs that could result from hypothermia, such as ECG changes, were carefully monitored. The esophageal temperature decreased from 33.0°C, which was measured immediately after entering the operation room, to 32.0°C at the time of cardiac arrest. The esophageal temperature further decreased to a minimum of 30.0°C during CPR. With rewarming, the temperature recovered to 33.0°C and finally, ROSC was achieved. Considering that chest compressions, drugs, and defibrillation did not induce ROSC, which was achieved with the recovery of an esophageal temperature of 33.0°C, the additional rapid fall in the patient’s core body temperature during anesthesia may have caused the cardiac arrest.
In this case study, we suspected that restenosis at the site of the previous stent insertion contributed to the cardiac arrest, given the following three findings: First, the patient had a normal preoperative ECG with no signs of ischemia but developed ST-segment depression during anesthesia. Second, repetitive cardiac arrests occurred even after achieving ROSC with rewarming. Third, an increased troponin-T concentration was observed, and suspected re-stenotic occlusion of the left main coronary artery was confirmed by Doppler TTE after the patient’s transfer to the ICU. Accordingly, myocardial infarction was considered the main cause of death.
The mechanisms of myocardial infarction owing to stent restenosis after DES insertion are rapid stent thrombosis or profuse in-stent restenosis (ISR).13 We assume that the lesion characteristics in the left main bifurcation of the coronary artery and DAPT interruption owing to the increasing amount of cerebral hemorrhage contributed to the stent restenosis in this patient. As evidence in support of this assumption, previous studies have shown that stenting at the left main bifurcation and DAPT interruption are strong risk factors for stent thrombosis or ISR.14,15 While controversial, not interrupting aspirin perioperatively could have avoided poor outcomes.16 However, there were no clinical signs related to expected ongoing restenosis of the coronary artery or myocardial ischemia preoperatively in our patient. We suspect that the possible ischemic signs may have been masked by the aforementioned cardioprotective benefits of therapeutic hypothermia. We also believe that the following three conditions contributed to the intraoperative myocardial ischemia and infarction caused by acute restenosis of the coronary stent: First, intraoperative hypotension may have contributed to myocardial injury. The IV anesthetics used during anesthesia, propofol and remifentanil, induced redistribution of blood volume by peripheral vasodilation, resulting in profound hypotension.4 Profound hypotension was strongly associated with myocardial injury and especially harmful for this patient, who had preexisting ischemic heart disease.17 Second, tachycardia, which may have been caused by the inotrope dopamine, used to compensate for the low BP or the stress response to surgical stimulation from shallow anesthetic depth, may have triggered myocardial ischemia by decreasing oxygen delivery and increasing oxygen consumption.17,18 If cardiogenic shock is suspected, as in our case, norepinephrine could be a better option than dopamine.19 Third, the bolus injection of phenylephrine, a selective α1-adrenergic agonist, just before cardiac arrest may have temporarily and abruptly increased cardiac afterload and impaired myocardial perfusion.20
Had we known that coronary artery occlusion would result from hypothermia-induced cardiac arrest and coronary stent restenosis, the treatment of choice would have been CABG surgery while maintaining circulation with emergency cardiopulmonary bypass (CPB) or extracorporeal membrane oxygenation (ECMO) via the femoral artery and vein.11 However, it was difficult to determine the etiology of the cardiac arrest at the time. Even if we had known the cause, ideal treatment would have been impossible, given circumstances beyond our control; all thoracic surgeons were busy operating, and emergency CPB and ECMO were not available. As an alternative, repeat PCI after the patient achieved ROSC could have been performed. However, the patient had a prolonged CPR duration, and he did not maintain the ROSC state for a period sufficient for intervention, which led to a poor outcome.
In this case, it was not possible to accurately determine the reason for the cardiac arrest and for not achieving consistent ROSC. Instead, we conclude that cardiac arrest developed and persisted as a result of the unintentional severe hypothermia that occurred during general anesthesia and/or coronary artery restenosis at the site of the previous PCI. A thorough evaluation of heart-related history is required for TBI patients who have a history of ischemic heart diseases and who undergo therapeutic hypothermia because there may be undiscovered coronary insufficiency owing to the cardioprotective effects of therapeutic hypothermia. In addition, careful monitoring of the patient’s core body temperature, and preventive and countermeasures should be fully prepared and taken against adverse hemodynamic effects, even cardiac arrest, which can occur because of additive hypothermic responses during anesthesia.
Ethics statement: This case report was approved by our Institutional Review Board (EMC 2020-05-005-001), and the case report was performed in compliance with the EQUATOR Network guidelines (the CARE guidelines). All patient details were de-identified, and written consent was obtained from the patient’s guardian for treatment and publication of this report.
Declaration of conflicting interest: The authors declare that there is no conflict of interest.
Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
ORCID iDs: Dong Ho Park https://orcid.org/0000-0002-6587-3756
Tae Woo Kim https://orcid.org/0000-0002-2115-6448
Mo Se Kim https://orcid.org/0000-0001-7362-0030 | Fatal | ReactionOutcome | CC BY-NC | 33499678 | 19,395,621 | 2021-01 |
What was the outcome of reaction 'Myocardial infarction'? | Cardiac arrest caused by accidental severe hypothermia and myocardial infarction during general anesthesia.
Therapeutic hypothermia is often used for traumatic brain injury because of its neuroprotective effect and decreased secondary brain injury. However, this procedure lacks clinical evidence supporting its efficacy, and adverse outcomes have been reported during general anesthesia. A 61-year-old man with a history of percutaneous coronary intervention (PCI) was admitted with traumatic brain injury. Immediately after admission, he underwent mild therapeutic hypothermia with a target temperature of 33.0°C for neuroprotection. During general anesthesia for emergency surgery because he developed a mass effect, hypothermic cardiac arrest occurred following an additional decrease in the core body temperature. Moreover, myocardial infarction caused by restenosis of the previous PCI lesion also contributed to the cardiac arrest. Although the patient recovered spontaneous circulation after an hour-long cardiopulmonary resuscitation with rewarming, he eventually died of subsequent repetitive cardiac arrests. When anesthetizing patients undergoing therapeutic hypothermia, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.
Introduction
Therapeutic hypothermia is an attractive strategy for traumatic brain injury (TBI), which demonstrates neuroprotective effects and decreases secondary brain injury.1 Despite the lack of clinical evidence regarding efficacy, therapeutic hypothermia after TBI has been used for at least half a century because of its potential benefits, which are supported by numerous preclinical studies.2,3 However, there have been no reports on the danger of unintentional severe hypothermia during general anesthesia caused by the additional drop in temperature of patients undergoing therapeutic hypothermia. When anesthetizing patients with comorbidities, such as ischemic heart disease, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.4,5
Case report
A 61-year-old man (height, 168 cm; weight, 70 kg) presented to our emergency room (ER) with head trauma after falling on a cement floor from a 5-meter-tall construction site. Upon arrival at the ER, the patient had a Glasgow coma scale score of 11 and the following vital signs: blood pressure (BP), 147/83 mmHg; heart rate (HR), 86 beats per minute (bpm); respiratory rate, 24 breaths per minute; body temperature from the auditory canal, 36.8°C; and percutaneous oxygen saturation (SpO2), 97%. Upon discovering signs of acute subdural hemorrhage, subarachnoid hemorrhage, and intraventricular hemorrhage from a brain computed tomography (CT) scan performed in the ER, we admitted the patient to the intensive care unit (ICU) for close monitoring. After approximately 6 hours, we performed follow-up brain CT because the patient’s mental status had deteriorated from drowsy to stupor. Upon observing a mass effect characterized by an increase in the hemorrhage volume and a left midline shift on the CT images, the patient was scheduled for immediate intubation and emergency decompressive craniectomy (Figure 1).
Figure 1. Brain computed tomography (CT) images taken 6 hours after intensive care unit admission. (a) Axial view showing acute subarachnoid hemorrhage in the peri-mesencephalic cistern (arrow A) and subdural hemorrhage in the right inferior temporal (arrow B) and right parietal (arrow C) regions. (b) Axial view showing acute subdural hemorrhage in the right temporal region (black arrow) and left midline shift.
Although the patient had a history of ischemic heart disease with percutaneous coronary intervention (PCI), it was difficult to obtain accurate information because there was no relevant data in the patient’s medical records from the authors’ hospital, and his drowsy state interfered with detailed interviews. According to the chest CT scan performed immediately after admission, coronary stents had been inserted at the bifurcation site of the left main coronary artery (Figure 2). It also appeared that the inserted stents were drug-eluting stents (DES) considering that he had received dual antiplatelet therapy (DAPT) with aspirin and clopidogrel (Plavix®; Sanofi, Bridgewater, NJ, USA) a switching platelet P2Y12 inhibitor, for approximately 4 months. However, DAPT was discontinued owing to the worsening cerebral hemorrhage after hospitalization; bridging antiplatelet therapy was also not performed. Before the surgery, the patient showed a normal electrocardiogram (ECG), and cardiac biomarker testing revealed a troponin-T concentration of 0.006 µg/L, which was within normal limits (Figure 3). The patient demonstrated satisfactory functional capacity to be able to work at a construction site just before the trauma. Therefore, an additional preoperative cardiac evaluation was not considered necessary.
Figure 2. Chest computed tomography (CT) images showing coronary stents in the main bifurcation of the left coronary arteries (white arrow in (a) and (b)).
Figure 3. Preoperative electrocardiogram of the patient showing normal sinus rhythm with a heart rate of 95 bpm.
bpm, beats per minute.
Immediately after admission to the ICU, the patient’s bladder temperature was 38.0°C. Using a hydrogel-coated water-circulating pad (Arctic Sun®; Medivance Inc., Louisville, CO, USA), which is a surface cooling device, the patient underwent mild therapeutic hypothermia with a target temperature of 33.0°C. During therapeutic hypothermia, sedation and shivering control were performed using continuous rate infusions of midazolam at 5 mg/hour and remifentanil at 0.2 µg/kg/minute. The patient’s bladder temperature was 33.4°C just before he was transferred to the operating room. Preoperative laboratory tests showed no unusual findings except for mildly elevated aspartate aminotransferase (73 U/L) and alanine aminotransferase (51 U/L) concentrations.
The patient’s vital signs upon arrival at the operating room were a BP of 130/70 mmHg and SpO2 of 100%. Lead II of the ECG was monitored, and there were no unusual findings except sinus tachycardia with an HR of 100 bpm on the initial ECG. Arterial BP (ABP), central venous pressure (CVP), and esophageal temperature were monitored from the right radial artery, right subclavian vein, and lower third of the esophagus using a temperature sensor attached to the esophageal stethoscope, respectively. The initial ABP, CVP, and esophageal temperature were 140/85 mmHg, 6 mmHg, and 33.0°C, respectively. Arterial blood gas analysis was performed in the arterial line, with values calculated as for a body temperature of 37°C, which revealed the following: pH, 7.400; partial pressure of carbon dioxide (PCO2), 40 mmHg; partial pressure of oxygen (PO2), 174 mmHg; bicarbonate (HCO3), 24.8 mmol/L; base excess, 0.0; lactate, 0.17 mmol/L; ionized calcium, 1.04 mmol/L; and potassium, 3.3 mmol/L. Total intravenous (IV) anesthesia was performed using a target-controlled infusion of propofol and remifentanil in accordance with the Schnider6 and Minto7 models, respectively. Propofol and remifentanil were infused continuously with effect-site concentrations (Ce) of 1.5 µg/mL and 1.5 ng/mL, respectively. Concurrently, 50 mg of rocuronium was administered by IV bolus injection for neuromuscular blockade. Mechanical ventilation was adjusted to maintain the end-tidal carbon dioxide value at 30 to 33 mmHg.
As ABP decreased to 78/42 mmHg and CVP decreased to 4 mmHg within 10 minutes after anesthesia, the Ce of propofol and remifentanil were lowered to 0.8 µg/mL and 1.0 ng/mL, respectively, and approximately 1 L of balanced crystalloid fluid loading was performed. However, the patient’s BP did not recover, and a dopamine continuous infusion (5 µg/kg/minute) was initiated. When a surgical incision was made 10 minutes after beginning the dopamine infusion, the patient’s ABP and HR increased to 115/70 mmHg and 120 bpm, respectively, and marked ST-segment depression (−2.4 mm) was found, which was not observed on the preoperative or initial ECGs (Figure 4). Subsequently, the depth of anesthesia was increased by increasing the Ce of propofol to 1.0 to 1.2 µg/mL, followed by a continuous infusion of isosorbide dinitrate (1 µg/kg/minute). While HR was maintained at 110 to 120 bpm, ABP gradually decreased to 70/40 mmHg after approximately 5 minutes. To treat the hypotension, isosorbide dinitrate was discontinued, and a bolus injection of 100 µg phenylephrine was administered, after which, ABP increased temporarily to 95/60 mmHg. Subsequently, HR decreased suddenly and rapidly, and ventricular fibrillation occurred. The anesthesiologists immediately discontinued the surgery and performed cardiopulmonary resuscitation (CPR). The patient’s esophageal temperature at the time was 32.0°C.
Figure 4. Intraoperative electrocardiogram of the patient showing sinus tachycardia with significant ST depression. The ST depression on the monitor represented approximately −2.4 at the time.
Chest compressions were performed with periodic defibrillation with 200 J, and epinephrine injections were administered every 3 minutes; these measures elicited no response. During CPR, ECG monitoring showed temporary severe bradycardia (HR, 10–30 bpm). However, this was considered pulseless electrical activity because no pulse was detected in the patient’s femoral artery. Soon afterward, there was a rapid transition to ventricular fibrillation, and CPR was repeated. At the time, the patient’s esophageal temperature dropped to 30.0°C. According to the refractory state after the defibrillation and drug injections, we considered that the cardiac arrest was caused by hypothermia. To reverse the cause, a forced-air warming device and a warming blanket (Bair Hugger®; Arizant Healthcare Inc., Eden Prairie, MN, USA) were used for surface warming, the IV fluids were warmed, and surgical wound closure and dressing were performed rapidly to prevent surface heat loss. The first arterial blood gas analysis after ventricular fibrillation, calculated as for a body temperature of 37°C, showed the following: pH, 7.690; PCO2, 15 mmHg; PO2, 260 mmHg; HCO3, 18.1 mmol/L; base excess, −1.2; lactate, 0.33 mmol/L; ionized calcium, 0.88 mmol/L; and potassium, 5.0 mmol/L. Follow-up arterial blood gas analyses were repeated periodically. As we continued to perform CPR, the patient developed severe metabolic acidosis and hypocalcemia characterized by a pH of 6.840, HCO3 of 15.4 mmol/L, and ionized calcium of 0.66 mmol/L. To correct the abnormal findings, we administered appropriate amounts of sodium bicarbonate and calcium chloride. After an hour-long CPR and rewarming, the patient’s esophageal temperature recovered to 33.0°C, which was followed by observing sinus rhythm characterized by an HR of 75 bpm on an ECG monitor and detecting a pulse in the femoral artery with a characteristic ABP waveform that eventually led to the recovery of BP to 130/55 mmHg. Upon confirming recovery of spontaneous circulation (ROSC), the patient was transported to the ICU while we administered dopamine (10 µg/kg/minute), norepinephrine (0.1 µg/kg/minute), and vasopressin (2 U/hour). ABP, HR, and esophageal temperature during anesthesia are shown in Figure 5.
Figure 5. Changes in vital signs (blood pressure (BP), heart rate (HR), and esophageal temperature) over time (5-minute intervals).
bpm, beats per minute.
Cardiac biomarker testing performed after the patient’s transfer to the ICU showed an increased troponin-T level of 1.432 µg/L. Portable Doppler transthoracic echocardiography (TTE) revealed total occlusion of the left main coronary artery, which had undergone PCI previously. Because emergency coronary artery bypass grafting (CABG) was impossible at the time, repeated PCI was considered an alternative treatment option. However, even after his transfer to the ICU, the patient experienced repetitive ventricular fibrillation. Four hours after the patient’s transfer to the ICU, the patient’s guardian requested that CPR should be stopped. Without resuscitation, the patient died.
Discussion
In this case, the patient, who was receiving therapeutic hypothermia, underwent emergency craniectomy owing to worsening brain hemorrhage, and cardiac arrest occurred during general anesthesia. We deduced that the cardiac arrest was owing to the additional decrease in the patient’s core body temperature and myocardial infarction caused by restenosis of the previous PCI-treated lesion.
Therapeutic hypothermia is used to improve neurological outcomes in cardiac arrest, TBI, stroke, and hypoxic–ischemic encephalopathy.1,4 Studies on the efficacy of therapeutic hypothermia for TBI have been less convincing, and they show conflicting results. However, hypothermia is consistently used as an option for TBI because it reduces intracranial pressure and the cerebral metabolic rate, thereby protecting neuronal function in the brain and decreasing the wide range of injuries or reparative processes by attenuating temperature-sensitive mechanisms, such as excitotoxicity, free radical generation, apoptosis, and inflammation.1,2
The cardiovascular effects associated with hypothermia are complex. Hypothermia affects the electrical conduction of myocardial cells, which can induce hypotension or bradycardia and even increase the risk of ventricular fibrillation by prolonging repolarization.4,8 However, mild hypothermia of 32.0°C to 35.0°C demonstrates cardioprotective effects by decreasing myocardial oxygen demand and increasing myocardial perfusion through vasodilation of coronary vessels.9 Therefore, there has been a growing interest in the clinical use of mild therapeutic hypothermia as an adjunctive therapy during ischemic heart disease.10 Our patient received mild therapeutic hypothermia with a target core temperature of 33.0°C, which may have yielded cardioprotective as well as neuroprotective benefits.11
During general anesthesia, hypothermia is induced following an increase in heat loss caused by peripheral vasodilation after the injection of anesthetic drugs, heat redistribution from the core to the periphery, and impaired normal thermoregulatory responses, such as vasoconstriction or shivering.5 In our patient, the cold operating room temperature and irrigating an uncovered surgical area in the brain with cold saline also may have contributed to the additional core body temperature drop in addition to the effects of general anesthesia.12 However, in our case, neurosurgeons requested that anesthesiologists not perform warming during anesthesia. Instead, esophageal temperature and signs that could result from hypothermia, such as ECG changes, were carefully monitored. The esophageal temperature decreased from 33.0°C, which was measured immediately after entering the operation room, to 32.0°C at the time of cardiac arrest. The esophageal temperature further decreased to a minimum of 30.0°C during CPR. With rewarming, the temperature recovered to 33.0°C and finally, ROSC was achieved. Considering that chest compressions, drugs, and defibrillation did not induce ROSC, which was achieved with the recovery of an esophageal temperature of 33.0°C, the additional rapid fall in the patient’s core body temperature during anesthesia may have caused the cardiac arrest.
In this case study, we suspected that restenosis at the site of the previous stent insertion contributed to the cardiac arrest, given the following three findings: First, the patient had a normal preoperative ECG with no signs of ischemia but developed ST-segment depression during anesthesia. Second, repetitive cardiac arrests occurred even after achieving ROSC with rewarming. Third, an increased troponin-T concentration was observed, and suspected re-stenotic occlusion of the left main coronary artery was confirmed by Doppler TTE after the patient’s transfer to the ICU. Accordingly, myocardial infarction was considered the main cause of death.
The mechanisms of myocardial infarction owing to stent restenosis after DES insertion are rapid stent thrombosis or profuse in-stent restenosis (ISR).13 We assume that the lesion characteristics in the left main bifurcation of the coronary artery and DAPT interruption owing to the increasing amount of cerebral hemorrhage contributed to the stent restenosis in this patient. As evidence in support of this assumption, previous studies have shown that stenting at the left main bifurcation and DAPT interruption are strong risk factors for stent thrombosis or ISR.14,15 While controversial, not interrupting aspirin perioperatively could have avoided poor outcomes.16 However, there were no clinical signs related to expected ongoing restenosis of the coronary artery or myocardial ischemia preoperatively in our patient. We suspect that the possible ischemic signs may have been masked by the aforementioned cardioprotective benefits of therapeutic hypothermia. We also believe that the following three conditions contributed to the intraoperative myocardial ischemia and infarction caused by acute restenosis of the coronary stent: First, intraoperative hypotension may have contributed to myocardial injury. The IV anesthetics used during anesthesia, propofol and remifentanil, induced redistribution of blood volume by peripheral vasodilation, resulting in profound hypotension.4 Profound hypotension was strongly associated with myocardial injury and especially harmful for this patient, who had preexisting ischemic heart disease.17 Second, tachycardia, which may have been caused by the inotrope dopamine, used to compensate for the low BP or the stress response to surgical stimulation from shallow anesthetic depth, may have triggered myocardial ischemia by decreasing oxygen delivery and increasing oxygen consumption.17,18 If cardiogenic shock is suspected, as in our case, norepinephrine could be a better option than dopamine.19 Third, the bolus injection of phenylephrine, a selective α1-adrenergic agonist, just before cardiac arrest may have temporarily and abruptly increased cardiac afterload and impaired myocardial perfusion.20
Had we known that coronary artery occlusion would result from hypothermia-induced cardiac arrest and coronary stent restenosis, the treatment of choice would have been CABG surgery while maintaining circulation with emergency cardiopulmonary bypass (CPB) or extracorporeal membrane oxygenation (ECMO) via the femoral artery and vein.11 However, it was difficult to determine the etiology of the cardiac arrest at the time. Even if we had known the cause, ideal treatment would have been impossible, given circumstances beyond our control; all thoracic surgeons were busy operating, and emergency CPB and ECMO were not available. As an alternative, repeat PCI after the patient achieved ROSC could have been performed. However, the patient had a prolonged CPR duration, and he did not maintain the ROSC state for a period sufficient for intervention, which led to a poor outcome.
In this case, it was not possible to accurately determine the reason for the cardiac arrest and for not achieving consistent ROSC. Instead, we conclude that cardiac arrest developed and persisted as a result of the unintentional severe hypothermia that occurred during general anesthesia and/or coronary artery restenosis at the site of the previous PCI. A thorough evaluation of heart-related history is required for TBI patients who have a history of ischemic heart diseases and who undergo therapeutic hypothermia because there may be undiscovered coronary insufficiency owing to the cardioprotective effects of therapeutic hypothermia. In addition, careful monitoring of the patient’s core body temperature, and preventive and countermeasures should be fully prepared and taken against adverse hemodynamic effects, even cardiac arrest, which can occur because of additive hypothermic responses during anesthesia.
Ethics statement: This case report was approved by our Institutional Review Board (EMC 2020-05-005-001), and the case report was performed in compliance with the EQUATOR Network guidelines (the CARE guidelines). All patient details were de-identified, and written consent was obtained from the patient’s guardian for treatment and publication of this report.
Declaration of conflicting interest: The authors declare that there is no conflict of interest.
Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
ORCID iDs: Dong Ho Park https://orcid.org/0000-0002-6587-3756
Tae Woo Kim https://orcid.org/0000-0002-2115-6448
Mo Se Kim https://orcid.org/0000-0001-7362-0030 | Fatal | ReactionOutcome | CC BY-NC | 33499678 | 19,429,456 | 2021-01 |
What was the outcome of reaction 'Myocardial ischaemia'? | Cardiac arrest caused by accidental severe hypothermia and myocardial infarction during general anesthesia.
Therapeutic hypothermia is often used for traumatic brain injury because of its neuroprotective effect and decreased secondary brain injury. However, this procedure lacks clinical evidence supporting its efficacy, and adverse outcomes have been reported during general anesthesia. A 61-year-old man with a history of percutaneous coronary intervention (PCI) was admitted with traumatic brain injury. Immediately after admission, he underwent mild therapeutic hypothermia with a target temperature of 33.0°C for neuroprotection. During general anesthesia for emergency surgery because he developed a mass effect, hypothermic cardiac arrest occurred following an additional decrease in the core body temperature. Moreover, myocardial infarction caused by restenosis of the previous PCI lesion also contributed to the cardiac arrest. Although the patient recovered spontaneous circulation after an hour-long cardiopulmonary resuscitation with rewarming, he eventually died of subsequent repetitive cardiac arrests. When anesthetizing patients undergoing therapeutic hypothermia, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.
Introduction
Therapeutic hypothermia is an attractive strategy for traumatic brain injury (TBI), which demonstrates neuroprotective effects and decreases secondary brain injury.1 Despite the lack of clinical evidence regarding efficacy, therapeutic hypothermia after TBI has been used for at least half a century because of its potential benefits, which are supported by numerous preclinical studies.2,3 However, there have been no reports on the danger of unintentional severe hypothermia during general anesthesia caused by the additional drop in temperature of patients undergoing therapeutic hypothermia. When anesthetizing patients with comorbidities, such as ischemic heart disease, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.4,5
Case report
A 61-year-old man (height, 168 cm; weight, 70 kg) presented to our emergency room (ER) with head trauma after falling on a cement floor from a 5-meter-tall construction site. Upon arrival at the ER, the patient had a Glasgow coma scale score of 11 and the following vital signs: blood pressure (BP), 147/83 mmHg; heart rate (HR), 86 beats per minute (bpm); respiratory rate, 24 breaths per minute; body temperature from the auditory canal, 36.8°C; and percutaneous oxygen saturation (SpO2), 97%. Upon discovering signs of acute subdural hemorrhage, subarachnoid hemorrhage, and intraventricular hemorrhage from a brain computed tomography (CT) scan performed in the ER, we admitted the patient to the intensive care unit (ICU) for close monitoring. After approximately 6 hours, we performed follow-up brain CT because the patient’s mental status had deteriorated from drowsy to stupor. Upon observing a mass effect characterized by an increase in the hemorrhage volume and a left midline shift on the CT images, the patient was scheduled for immediate intubation and emergency decompressive craniectomy (Figure 1).
Figure 1. Brain computed tomography (CT) images taken 6 hours after intensive care unit admission. (a) Axial view showing acute subarachnoid hemorrhage in the peri-mesencephalic cistern (arrow A) and subdural hemorrhage in the right inferior temporal (arrow B) and right parietal (arrow C) regions. (b) Axial view showing acute subdural hemorrhage in the right temporal region (black arrow) and left midline shift.
Although the patient had a history of ischemic heart disease with percutaneous coronary intervention (PCI), it was difficult to obtain accurate information because there was no relevant data in the patient’s medical records from the authors’ hospital, and his drowsy state interfered with detailed interviews. According to the chest CT scan performed immediately after admission, coronary stents had been inserted at the bifurcation site of the left main coronary artery (Figure 2). It also appeared that the inserted stents were drug-eluting stents (DES) considering that he had received dual antiplatelet therapy (DAPT) with aspirin and clopidogrel (Plavix®; Sanofi, Bridgewater, NJ, USA) a switching platelet P2Y12 inhibitor, for approximately 4 months. However, DAPT was discontinued owing to the worsening cerebral hemorrhage after hospitalization; bridging antiplatelet therapy was also not performed. Before the surgery, the patient showed a normal electrocardiogram (ECG), and cardiac biomarker testing revealed a troponin-T concentration of 0.006 µg/L, which was within normal limits (Figure 3). The patient demonstrated satisfactory functional capacity to be able to work at a construction site just before the trauma. Therefore, an additional preoperative cardiac evaluation was not considered necessary.
Figure 2. Chest computed tomography (CT) images showing coronary stents in the main bifurcation of the left coronary arteries (white arrow in (a) and (b)).
Figure 3. Preoperative electrocardiogram of the patient showing normal sinus rhythm with a heart rate of 95 bpm.
bpm, beats per minute.
Immediately after admission to the ICU, the patient’s bladder temperature was 38.0°C. Using a hydrogel-coated water-circulating pad (Arctic Sun®; Medivance Inc., Louisville, CO, USA), which is a surface cooling device, the patient underwent mild therapeutic hypothermia with a target temperature of 33.0°C. During therapeutic hypothermia, sedation and shivering control were performed using continuous rate infusions of midazolam at 5 mg/hour and remifentanil at 0.2 µg/kg/minute. The patient’s bladder temperature was 33.4°C just before he was transferred to the operating room. Preoperative laboratory tests showed no unusual findings except for mildly elevated aspartate aminotransferase (73 U/L) and alanine aminotransferase (51 U/L) concentrations.
The patient’s vital signs upon arrival at the operating room were a BP of 130/70 mmHg and SpO2 of 100%. Lead II of the ECG was monitored, and there were no unusual findings except sinus tachycardia with an HR of 100 bpm on the initial ECG. Arterial BP (ABP), central venous pressure (CVP), and esophageal temperature were monitored from the right radial artery, right subclavian vein, and lower third of the esophagus using a temperature sensor attached to the esophageal stethoscope, respectively. The initial ABP, CVP, and esophageal temperature were 140/85 mmHg, 6 mmHg, and 33.0°C, respectively. Arterial blood gas analysis was performed in the arterial line, with values calculated as for a body temperature of 37°C, which revealed the following: pH, 7.400; partial pressure of carbon dioxide (PCO2), 40 mmHg; partial pressure of oxygen (PO2), 174 mmHg; bicarbonate (HCO3), 24.8 mmol/L; base excess, 0.0; lactate, 0.17 mmol/L; ionized calcium, 1.04 mmol/L; and potassium, 3.3 mmol/L. Total intravenous (IV) anesthesia was performed using a target-controlled infusion of propofol and remifentanil in accordance with the Schnider6 and Minto7 models, respectively. Propofol and remifentanil were infused continuously with effect-site concentrations (Ce) of 1.5 µg/mL and 1.5 ng/mL, respectively. Concurrently, 50 mg of rocuronium was administered by IV bolus injection for neuromuscular blockade. Mechanical ventilation was adjusted to maintain the end-tidal carbon dioxide value at 30 to 33 mmHg.
As ABP decreased to 78/42 mmHg and CVP decreased to 4 mmHg within 10 minutes after anesthesia, the Ce of propofol and remifentanil were lowered to 0.8 µg/mL and 1.0 ng/mL, respectively, and approximately 1 L of balanced crystalloid fluid loading was performed. However, the patient’s BP did not recover, and a dopamine continuous infusion (5 µg/kg/minute) was initiated. When a surgical incision was made 10 minutes after beginning the dopamine infusion, the patient’s ABP and HR increased to 115/70 mmHg and 120 bpm, respectively, and marked ST-segment depression (−2.4 mm) was found, which was not observed on the preoperative or initial ECGs (Figure 4). Subsequently, the depth of anesthesia was increased by increasing the Ce of propofol to 1.0 to 1.2 µg/mL, followed by a continuous infusion of isosorbide dinitrate (1 µg/kg/minute). While HR was maintained at 110 to 120 bpm, ABP gradually decreased to 70/40 mmHg after approximately 5 minutes. To treat the hypotension, isosorbide dinitrate was discontinued, and a bolus injection of 100 µg phenylephrine was administered, after which, ABP increased temporarily to 95/60 mmHg. Subsequently, HR decreased suddenly and rapidly, and ventricular fibrillation occurred. The anesthesiologists immediately discontinued the surgery and performed cardiopulmonary resuscitation (CPR). The patient’s esophageal temperature at the time was 32.0°C.
Figure 4. Intraoperative electrocardiogram of the patient showing sinus tachycardia with significant ST depression. The ST depression on the monitor represented approximately −2.4 at the time.
Chest compressions were performed with periodic defibrillation with 200 J, and epinephrine injections were administered every 3 minutes; these measures elicited no response. During CPR, ECG monitoring showed temporary severe bradycardia (HR, 10–30 bpm). However, this was considered pulseless electrical activity because no pulse was detected in the patient’s femoral artery. Soon afterward, there was a rapid transition to ventricular fibrillation, and CPR was repeated. At the time, the patient’s esophageal temperature dropped to 30.0°C. According to the refractory state after the defibrillation and drug injections, we considered that the cardiac arrest was caused by hypothermia. To reverse the cause, a forced-air warming device and a warming blanket (Bair Hugger®; Arizant Healthcare Inc., Eden Prairie, MN, USA) were used for surface warming, the IV fluids were warmed, and surgical wound closure and dressing were performed rapidly to prevent surface heat loss. The first arterial blood gas analysis after ventricular fibrillation, calculated as for a body temperature of 37°C, showed the following: pH, 7.690; PCO2, 15 mmHg; PO2, 260 mmHg; HCO3, 18.1 mmol/L; base excess, −1.2; lactate, 0.33 mmol/L; ionized calcium, 0.88 mmol/L; and potassium, 5.0 mmol/L. Follow-up arterial blood gas analyses were repeated periodically. As we continued to perform CPR, the patient developed severe metabolic acidosis and hypocalcemia characterized by a pH of 6.840, HCO3 of 15.4 mmol/L, and ionized calcium of 0.66 mmol/L. To correct the abnormal findings, we administered appropriate amounts of sodium bicarbonate and calcium chloride. After an hour-long CPR and rewarming, the patient’s esophageal temperature recovered to 33.0°C, which was followed by observing sinus rhythm characterized by an HR of 75 bpm on an ECG monitor and detecting a pulse in the femoral artery with a characteristic ABP waveform that eventually led to the recovery of BP to 130/55 mmHg. Upon confirming recovery of spontaneous circulation (ROSC), the patient was transported to the ICU while we administered dopamine (10 µg/kg/minute), norepinephrine (0.1 µg/kg/minute), and vasopressin (2 U/hour). ABP, HR, and esophageal temperature during anesthesia are shown in Figure 5.
Figure 5. Changes in vital signs (blood pressure (BP), heart rate (HR), and esophageal temperature) over time (5-minute intervals).
bpm, beats per minute.
Cardiac biomarker testing performed after the patient’s transfer to the ICU showed an increased troponin-T level of 1.432 µg/L. Portable Doppler transthoracic echocardiography (TTE) revealed total occlusion of the left main coronary artery, which had undergone PCI previously. Because emergency coronary artery bypass grafting (CABG) was impossible at the time, repeated PCI was considered an alternative treatment option. However, even after his transfer to the ICU, the patient experienced repetitive ventricular fibrillation. Four hours after the patient’s transfer to the ICU, the patient’s guardian requested that CPR should be stopped. Without resuscitation, the patient died.
Discussion
In this case, the patient, who was receiving therapeutic hypothermia, underwent emergency craniectomy owing to worsening brain hemorrhage, and cardiac arrest occurred during general anesthesia. We deduced that the cardiac arrest was owing to the additional decrease in the patient’s core body temperature and myocardial infarction caused by restenosis of the previous PCI-treated lesion.
Therapeutic hypothermia is used to improve neurological outcomes in cardiac arrest, TBI, stroke, and hypoxic–ischemic encephalopathy.1,4 Studies on the efficacy of therapeutic hypothermia for TBI have been less convincing, and they show conflicting results. However, hypothermia is consistently used as an option for TBI because it reduces intracranial pressure and the cerebral metabolic rate, thereby protecting neuronal function in the brain and decreasing the wide range of injuries or reparative processes by attenuating temperature-sensitive mechanisms, such as excitotoxicity, free radical generation, apoptosis, and inflammation.1,2
The cardiovascular effects associated with hypothermia are complex. Hypothermia affects the electrical conduction of myocardial cells, which can induce hypotension or bradycardia and even increase the risk of ventricular fibrillation by prolonging repolarization.4,8 However, mild hypothermia of 32.0°C to 35.0°C demonstrates cardioprotective effects by decreasing myocardial oxygen demand and increasing myocardial perfusion through vasodilation of coronary vessels.9 Therefore, there has been a growing interest in the clinical use of mild therapeutic hypothermia as an adjunctive therapy during ischemic heart disease.10 Our patient received mild therapeutic hypothermia with a target core temperature of 33.0°C, which may have yielded cardioprotective as well as neuroprotective benefits.11
During general anesthesia, hypothermia is induced following an increase in heat loss caused by peripheral vasodilation after the injection of anesthetic drugs, heat redistribution from the core to the periphery, and impaired normal thermoregulatory responses, such as vasoconstriction or shivering.5 In our patient, the cold operating room temperature and irrigating an uncovered surgical area in the brain with cold saline also may have contributed to the additional core body temperature drop in addition to the effects of general anesthesia.12 However, in our case, neurosurgeons requested that anesthesiologists not perform warming during anesthesia. Instead, esophageal temperature and signs that could result from hypothermia, such as ECG changes, were carefully monitored. The esophageal temperature decreased from 33.0°C, which was measured immediately after entering the operation room, to 32.0°C at the time of cardiac arrest. The esophageal temperature further decreased to a minimum of 30.0°C during CPR. With rewarming, the temperature recovered to 33.0°C and finally, ROSC was achieved. Considering that chest compressions, drugs, and defibrillation did not induce ROSC, which was achieved with the recovery of an esophageal temperature of 33.0°C, the additional rapid fall in the patient’s core body temperature during anesthesia may have caused the cardiac arrest.
In this case study, we suspected that restenosis at the site of the previous stent insertion contributed to the cardiac arrest, given the following three findings: First, the patient had a normal preoperative ECG with no signs of ischemia but developed ST-segment depression during anesthesia. Second, repetitive cardiac arrests occurred even after achieving ROSC with rewarming. Third, an increased troponin-T concentration was observed, and suspected re-stenotic occlusion of the left main coronary artery was confirmed by Doppler TTE after the patient’s transfer to the ICU. Accordingly, myocardial infarction was considered the main cause of death.
The mechanisms of myocardial infarction owing to stent restenosis after DES insertion are rapid stent thrombosis or profuse in-stent restenosis (ISR).13 We assume that the lesion characteristics in the left main bifurcation of the coronary artery and DAPT interruption owing to the increasing amount of cerebral hemorrhage contributed to the stent restenosis in this patient. As evidence in support of this assumption, previous studies have shown that stenting at the left main bifurcation and DAPT interruption are strong risk factors for stent thrombosis or ISR.14,15 While controversial, not interrupting aspirin perioperatively could have avoided poor outcomes.16 However, there were no clinical signs related to expected ongoing restenosis of the coronary artery or myocardial ischemia preoperatively in our patient. We suspect that the possible ischemic signs may have been masked by the aforementioned cardioprotective benefits of therapeutic hypothermia. We also believe that the following three conditions contributed to the intraoperative myocardial ischemia and infarction caused by acute restenosis of the coronary stent: First, intraoperative hypotension may have contributed to myocardial injury. The IV anesthetics used during anesthesia, propofol and remifentanil, induced redistribution of blood volume by peripheral vasodilation, resulting in profound hypotension.4 Profound hypotension was strongly associated with myocardial injury and especially harmful for this patient, who had preexisting ischemic heart disease.17 Second, tachycardia, which may have been caused by the inotrope dopamine, used to compensate for the low BP or the stress response to surgical stimulation from shallow anesthetic depth, may have triggered myocardial ischemia by decreasing oxygen delivery and increasing oxygen consumption.17,18 If cardiogenic shock is suspected, as in our case, norepinephrine could be a better option than dopamine.19 Third, the bolus injection of phenylephrine, a selective α1-adrenergic agonist, just before cardiac arrest may have temporarily and abruptly increased cardiac afterload and impaired myocardial perfusion.20
Had we known that coronary artery occlusion would result from hypothermia-induced cardiac arrest and coronary stent restenosis, the treatment of choice would have been CABG surgery while maintaining circulation with emergency cardiopulmonary bypass (CPB) or extracorporeal membrane oxygenation (ECMO) via the femoral artery and vein.11 However, it was difficult to determine the etiology of the cardiac arrest at the time. Even if we had known the cause, ideal treatment would have been impossible, given circumstances beyond our control; all thoracic surgeons were busy operating, and emergency CPB and ECMO were not available. As an alternative, repeat PCI after the patient achieved ROSC could have been performed. However, the patient had a prolonged CPR duration, and he did not maintain the ROSC state for a period sufficient for intervention, which led to a poor outcome.
In this case, it was not possible to accurately determine the reason for the cardiac arrest and for not achieving consistent ROSC. Instead, we conclude that cardiac arrest developed and persisted as a result of the unintentional severe hypothermia that occurred during general anesthesia and/or coronary artery restenosis at the site of the previous PCI. A thorough evaluation of heart-related history is required for TBI patients who have a history of ischemic heart diseases and who undergo therapeutic hypothermia because there may be undiscovered coronary insufficiency owing to the cardioprotective effects of therapeutic hypothermia. In addition, careful monitoring of the patient’s core body temperature, and preventive and countermeasures should be fully prepared and taken against adverse hemodynamic effects, even cardiac arrest, which can occur because of additive hypothermic responses during anesthesia.
Ethics statement: This case report was approved by our Institutional Review Board (EMC 2020-05-005-001), and the case report was performed in compliance with the EQUATOR Network guidelines (the CARE guidelines). All patient details were de-identified, and written consent was obtained from the patient’s guardian for treatment and publication of this report.
Declaration of conflicting interest: The authors declare that there is no conflict of interest.
Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
ORCID iDs: Dong Ho Park https://orcid.org/0000-0002-6587-3756
Tae Woo Kim https://orcid.org/0000-0002-2115-6448
Mo Se Kim https://orcid.org/0000-0001-7362-0030 | Fatal | ReactionOutcome | CC BY-NC | 33499678 | 19,429,456 | 2021-01 |
What was the outcome of reaction 'Ventricular fibrillation'? | Cardiac arrest caused by accidental severe hypothermia and myocardial infarction during general anesthesia.
Therapeutic hypothermia is often used for traumatic brain injury because of its neuroprotective effect and decreased secondary brain injury. However, this procedure lacks clinical evidence supporting its efficacy, and adverse outcomes have been reported during general anesthesia. A 61-year-old man with a history of percutaneous coronary intervention (PCI) was admitted with traumatic brain injury. Immediately after admission, he underwent mild therapeutic hypothermia with a target temperature of 33.0°C for neuroprotection. During general anesthesia for emergency surgery because he developed a mass effect, hypothermic cardiac arrest occurred following an additional decrease in the core body temperature. Moreover, myocardial infarction caused by restenosis of the previous PCI lesion also contributed to the cardiac arrest. Although the patient recovered spontaneous circulation after an hour-long cardiopulmonary resuscitation with rewarming, he eventually died of subsequent repetitive cardiac arrests. When anesthetizing patients undergoing therapeutic hypothermia, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.
Introduction
Therapeutic hypothermia is an attractive strategy for traumatic brain injury (TBI), which demonstrates neuroprotective effects and decreases secondary brain injury.1 Despite the lack of clinical evidence regarding efficacy, therapeutic hypothermia after TBI has been used for at least half a century because of its potential benefits, which are supported by numerous preclinical studies.2,3 However, there have been no reports on the danger of unintentional severe hypothermia during general anesthesia caused by the additional drop in temperature of patients undergoing therapeutic hypothermia. When anesthetizing patients with comorbidities, such as ischemic heart disease, caution is required to prevent adverse outcomes that can be caused by unintentional severe hypothermia and exacerbation of underlying heart disease.4,5
Case report
A 61-year-old man (height, 168 cm; weight, 70 kg) presented to our emergency room (ER) with head trauma after falling on a cement floor from a 5-meter-tall construction site. Upon arrival at the ER, the patient had a Glasgow coma scale score of 11 and the following vital signs: blood pressure (BP), 147/83 mmHg; heart rate (HR), 86 beats per minute (bpm); respiratory rate, 24 breaths per minute; body temperature from the auditory canal, 36.8°C; and percutaneous oxygen saturation (SpO2), 97%. Upon discovering signs of acute subdural hemorrhage, subarachnoid hemorrhage, and intraventricular hemorrhage from a brain computed tomography (CT) scan performed in the ER, we admitted the patient to the intensive care unit (ICU) for close monitoring. After approximately 6 hours, we performed follow-up brain CT because the patient’s mental status had deteriorated from drowsy to stupor. Upon observing a mass effect characterized by an increase in the hemorrhage volume and a left midline shift on the CT images, the patient was scheduled for immediate intubation and emergency decompressive craniectomy (Figure 1).
Figure 1. Brain computed tomography (CT) images taken 6 hours after intensive care unit admission. (a) Axial view showing acute subarachnoid hemorrhage in the peri-mesencephalic cistern (arrow A) and subdural hemorrhage in the right inferior temporal (arrow B) and right parietal (arrow C) regions. (b) Axial view showing acute subdural hemorrhage in the right temporal region (black arrow) and left midline shift.
Although the patient had a history of ischemic heart disease with percutaneous coronary intervention (PCI), it was difficult to obtain accurate information because there was no relevant data in the patient’s medical records from the authors’ hospital, and his drowsy state interfered with detailed interviews. According to the chest CT scan performed immediately after admission, coronary stents had been inserted at the bifurcation site of the left main coronary artery (Figure 2). It also appeared that the inserted stents were drug-eluting stents (DES) considering that he had received dual antiplatelet therapy (DAPT) with aspirin and clopidogrel (Plavix®; Sanofi, Bridgewater, NJ, USA) a switching platelet P2Y12 inhibitor, for approximately 4 months. However, DAPT was discontinued owing to the worsening cerebral hemorrhage after hospitalization; bridging antiplatelet therapy was also not performed. Before the surgery, the patient showed a normal electrocardiogram (ECG), and cardiac biomarker testing revealed a troponin-T concentration of 0.006 µg/L, which was within normal limits (Figure 3). The patient demonstrated satisfactory functional capacity to be able to work at a construction site just before the trauma. Therefore, an additional preoperative cardiac evaluation was not considered necessary.
Figure 2. Chest computed tomography (CT) images showing coronary stents in the main bifurcation of the left coronary arteries (white arrow in (a) and (b)).
Figure 3. Preoperative electrocardiogram of the patient showing normal sinus rhythm with a heart rate of 95 bpm.
bpm, beats per minute.
Immediately after admission to the ICU, the patient’s bladder temperature was 38.0°C. Using a hydrogel-coated water-circulating pad (Arctic Sun®; Medivance Inc., Louisville, CO, USA), which is a surface cooling device, the patient underwent mild therapeutic hypothermia with a target temperature of 33.0°C. During therapeutic hypothermia, sedation and shivering control were performed using continuous rate infusions of midazolam at 5 mg/hour and remifentanil at 0.2 µg/kg/minute. The patient’s bladder temperature was 33.4°C just before he was transferred to the operating room. Preoperative laboratory tests showed no unusual findings except for mildly elevated aspartate aminotransferase (73 U/L) and alanine aminotransferase (51 U/L) concentrations.
The patient’s vital signs upon arrival at the operating room were a BP of 130/70 mmHg and SpO2 of 100%. Lead II of the ECG was monitored, and there were no unusual findings except sinus tachycardia with an HR of 100 bpm on the initial ECG. Arterial BP (ABP), central venous pressure (CVP), and esophageal temperature were monitored from the right radial artery, right subclavian vein, and lower third of the esophagus using a temperature sensor attached to the esophageal stethoscope, respectively. The initial ABP, CVP, and esophageal temperature were 140/85 mmHg, 6 mmHg, and 33.0°C, respectively. Arterial blood gas analysis was performed in the arterial line, with values calculated as for a body temperature of 37°C, which revealed the following: pH, 7.400; partial pressure of carbon dioxide (PCO2), 40 mmHg; partial pressure of oxygen (PO2), 174 mmHg; bicarbonate (HCO3), 24.8 mmol/L; base excess, 0.0; lactate, 0.17 mmol/L; ionized calcium, 1.04 mmol/L; and potassium, 3.3 mmol/L. Total intravenous (IV) anesthesia was performed using a target-controlled infusion of propofol and remifentanil in accordance with the Schnider6 and Minto7 models, respectively. Propofol and remifentanil were infused continuously with effect-site concentrations (Ce) of 1.5 µg/mL and 1.5 ng/mL, respectively. Concurrently, 50 mg of rocuronium was administered by IV bolus injection for neuromuscular blockade. Mechanical ventilation was adjusted to maintain the end-tidal carbon dioxide value at 30 to 33 mmHg.
As ABP decreased to 78/42 mmHg and CVP decreased to 4 mmHg within 10 minutes after anesthesia, the Ce of propofol and remifentanil were lowered to 0.8 µg/mL and 1.0 ng/mL, respectively, and approximately 1 L of balanced crystalloid fluid loading was performed. However, the patient’s BP did not recover, and a dopamine continuous infusion (5 µg/kg/minute) was initiated. When a surgical incision was made 10 minutes after beginning the dopamine infusion, the patient’s ABP and HR increased to 115/70 mmHg and 120 bpm, respectively, and marked ST-segment depression (−2.4 mm) was found, which was not observed on the preoperative or initial ECGs (Figure 4). Subsequently, the depth of anesthesia was increased by increasing the Ce of propofol to 1.0 to 1.2 µg/mL, followed by a continuous infusion of isosorbide dinitrate (1 µg/kg/minute). While HR was maintained at 110 to 120 bpm, ABP gradually decreased to 70/40 mmHg after approximately 5 minutes. To treat the hypotension, isosorbide dinitrate was discontinued, and a bolus injection of 100 µg phenylephrine was administered, after which, ABP increased temporarily to 95/60 mmHg. Subsequently, HR decreased suddenly and rapidly, and ventricular fibrillation occurred. The anesthesiologists immediately discontinued the surgery and performed cardiopulmonary resuscitation (CPR). The patient’s esophageal temperature at the time was 32.0°C.
Figure 4. Intraoperative electrocardiogram of the patient showing sinus tachycardia with significant ST depression. The ST depression on the monitor represented approximately −2.4 at the time.
Chest compressions were performed with periodic defibrillation with 200 J, and epinephrine injections were administered every 3 minutes; these measures elicited no response. During CPR, ECG monitoring showed temporary severe bradycardia (HR, 10–30 bpm). However, this was considered pulseless electrical activity because no pulse was detected in the patient’s femoral artery. Soon afterward, there was a rapid transition to ventricular fibrillation, and CPR was repeated. At the time, the patient’s esophageal temperature dropped to 30.0°C. According to the refractory state after the defibrillation and drug injections, we considered that the cardiac arrest was caused by hypothermia. To reverse the cause, a forced-air warming device and a warming blanket (Bair Hugger®; Arizant Healthcare Inc., Eden Prairie, MN, USA) were used for surface warming, the IV fluids were warmed, and surgical wound closure and dressing were performed rapidly to prevent surface heat loss. The first arterial blood gas analysis after ventricular fibrillation, calculated as for a body temperature of 37°C, showed the following: pH, 7.690; PCO2, 15 mmHg; PO2, 260 mmHg; HCO3, 18.1 mmol/L; base excess, −1.2; lactate, 0.33 mmol/L; ionized calcium, 0.88 mmol/L; and potassium, 5.0 mmol/L. Follow-up arterial blood gas analyses were repeated periodically. As we continued to perform CPR, the patient developed severe metabolic acidosis and hypocalcemia characterized by a pH of 6.840, HCO3 of 15.4 mmol/L, and ionized calcium of 0.66 mmol/L. To correct the abnormal findings, we administered appropriate amounts of sodium bicarbonate and calcium chloride. After an hour-long CPR and rewarming, the patient’s esophageal temperature recovered to 33.0°C, which was followed by observing sinus rhythm characterized by an HR of 75 bpm on an ECG monitor and detecting a pulse in the femoral artery with a characteristic ABP waveform that eventually led to the recovery of BP to 130/55 mmHg. Upon confirming recovery of spontaneous circulation (ROSC), the patient was transported to the ICU while we administered dopamine (10 µg/kg/minute), norepinephrine (0.1 µg/kg/minute), and vasopressin (2 U/hour). ABP, HR, and esophageal temperature during anesthesia are shown in Figure 5.
Figure 5. Changes in vital signs (blood pressure (BP), heart rate (HR), and esophageal temperature) over time (5-minute intervals).
bpm, beats per minute.
Cardiac biomarker testing performed after the patient’s transfer to the ICU showed an increased troponin-T level of 1.432 µg/L. Portable Doppler transthoracic echocardiography (TTE) revealed total occlusion of the left main coronary artery, which had undergone PCI previously. Because emergency coronary artery bypass grafting (CABG) was impossible at the time, repeated PCI was considered an alternative treatment option. However, even after his transfer to the ICU, the patient experienced repetitive ventricular fibrillation. Four hours after the patient’s transfer to the ICU, the patient’s guardian requested that CPR should be stopped. Without resuscitation, the patient died.
Discussion
In this case, the patient, who was receiving therapeutic hypothermia, underwent emergency craniectomy owing to worsening brain hemorrhage, and cardiac arrest occurred during general anesthesia. We deduced that the cardiac arrest was owing to the additional decrease in the patient’s core body temperature and myocardial infarction caused by restenosis of the previous PCI-treated lesion.
Therapeutic hypothermia is used to improve neurological outcomes in cardiac arrest, TBI, stroke, and hypoxic–ischemic encephalopathy.1,4 Studies on the efficacy of therapeutic hypothermia for TBI have been less convincing, and they show conflicting results. However, hypothermia is consistently used as an option for TBI because it reduces intracranial pressure and the cerebral metabolic rate, thereby protecting neuronal function in the brain and decreasing the wide range of injuries or reparative processes by attenuating temperature-sensitive mechanisms, such as excitotoxicity, free radical generation, apoptosis, and inflammation.1,2
The cardiovascular effects associated with hypothermia are complex. Hypothermia affects the electrical conduction of myocardial cells, which can induce hypotension or bradycardia and even increase the risk of ventricular fibrillation by prolonging repolarization.4,8 However, mild hypothermia of 32.0°C to 35.0°C demonstrates cardioprotective effects by decreasing myocardial oxygen demand and increasing myocardial perfusion through vasodilation of coronary vessels.9 Therefore, there has been a growing interest in the clinical use of mild therapeutic hypothermia as an adjunctive therapy during ischemic heart disease.10 Our patient received mild therapeutic hypothermia with a target core temperature of 33.0°C, which may have yielded cardioprotective as well as neuroprotective benefits.11
During general anesthesia, hypothermia is induced following an increase in heat loss caused by peripheral vasodilation after the injection of anesthetic drugs, heat redistribution from the core to the periphery, and impaired normal thermoregulatory responses, such as vasoconstriction or shivering.5 In our patient, the cold operating room temperature and irrigating an uncovered surgical area in the brain with cold saline also may have contributed to the additional core body temperature drop in addition to the effects of general anesthesia.12 However, in our case, neurosurgeons requested that anesthesiologists not perform warming during anesthesia. Instead, esophageal temperature and signs that could result from hypothermia, such as ECG changes, were carefully monitored. The esophageal temperature decreased from 33.0°C, which was measured immediately after entering the operation room, to 32.0°C at the time of cardiac arrest. The esophageal temperature further decreased to a minimum of 30.0°C during CPR. With rewarming, the temperature recovered to 33.0°C and finally, ROSC was achieved. Considering that chest compressions, drugs, and defibrillation did not induce ROSC, which was achieved with the recovery of an esophageal temperature of 33.0°C, the additional rapid fall in the patient’s core body temperature during anesthesia may have caused the cardiac arrest.
In this case study, we suspected that restenosis at the site of the previous stent insertion contributed to the cardiac arrest, given the following three findings: First, the patient had a normal preoperative ECG with no signs of ischemia but developed ST-segment depression during anesthesia. Second, repetitive cardiac arrests occurred even after achieving ROSC with rewarming. Third, an increased troponin-T concentration was observed, and suspected re-stenotic occlusion of the left main coronary artery was confirmed by Doppler TTE after the patient’s transfer to the ICU. Accordingly, myocardial infarction was considered the main cause of death.
The mechanisms of myocardial infarction owing to stent restenosis after DES insertion are rapid stent thrombosis or profuse in-stent restenosis (ISR).13 We assume that the lesion characteristics in the left main bifurcation of the coronary artery and DAPT interruption owing to the increasing amount of cerebral hemorrhage contributed to the stent restenosis in this patient. As evidence in support of this assumption, previous studies have shown that stenting at the left main bifurcation and DAPT interruption are strong risk factors for stent thrombosis or ISR.14,15 While controversial, not interrupting aspirin perioperatively could have avoided poor outcomes.16 However, there were no clinical signs related to expected ongoing restenosis of the coronary artery or myocardial ischemia preoperatively in our patient. We suspect that the possible ischemic signs may have been masked by the aforementioned cardioprotective benefits of therapeutic hypothermia. We also believe that the following three conditions contributed to the intraoperative myocardial ischemia and infarction caused by acute restenosis of the coronary stent: First, intraoperative hypotension may have contributed to myocardial injury. The IV anesthetics used during anesthesia, propofol and remifentanil, induced redistribution of blood volume by peripheral vasodilation, resulting in profound hypotension.4 Profound hypotension was strongly associated with myocardial injury and especially harmful for this patient, who had preexisting ischemic heart disease.17 Second, tachycardia, which may have been caused by the inotrope dopamine, used to compensate for the low BP or the stress response to surgical stimulation from shallow anesthetic depth, may have triggered myocardial ischemia by decreasing oxygen delivery and increasing oxygen consumption.17,18 If cardiogenic shock is suspected, as in our case, norepinephrine could be a better option than dopamine.19 Third, the bolus injection of phenylephrine, a selective α1-adrenergic agonist, just before cardiac arrest may have temporarily and abruptly increased cardiac afterload and impaired myocardial perfusion.20
Had we known that coronary artery occlusion would result from hypothermia-induced cardiac arrest and coronary stent restenosis, the treatment of choice would have been CABG surgery while maintaining circulation with emergency cardiopulmonary bypass (CPB) or extracorporeal membrane oxygenation (ECMO) via the femoral artery and vein.11 However, it was difficult to determine the etiology of the cardiac arrest at the time. Even if we had known the cause, ideal treatment would have been impossible, given circumstances beyond our control; all thoracic surgeons were busy operating, and emergency CPB and ECMO were not available. As an alternative, repeat PCI after the patient achieved ROSC could have been performed. However, the patient had a prolonged CPR duration, and he did not maintain the ROSC state for a period sufficient for intervention, which led to a poor outcome.
In this case, it was not possible to accurately determine the reason for the cardiac arrest and for not achieving consistent ROSC. Instead, we conclude that cardiac arrest developed and persisted as a result of the unintentional severe hypothermia that occurred during general anesthesia and/or coronary artery restenosis at the site of the previous PCI. A thorough evaluation of heart-related history is required for TBI patients who have a history of ischemic heart diseases and who undergo therapeutic hypothermia because there may be undiscovered coronary insufficiency owing to the cardioprotective effects of therapeutic hypothermia. In addition, careful monitoring of the patient’s core body temperature, and preventive and countermeasures should be fully prepared and taken against adverse hemodynamic effects, even cardiac arrest, which can occur because of additive hypothermic responses during anesthesia.
Ethics statement: This case report was approved by our Institutional Review Board (EMC 2020-05-005-001), and the case report was performed in compliance with the EQUATOR Network guidelines (the CARE guidelines). All patient details were de-identified, and written consent was obtained from the patient’s guardian for treatment and publication of this report.
Declaration of conflicting interest: The authors declare that there is no conflict of interest.
Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
ORCID iDs: Dong Ho Park https://orcid.org/0000-0002-6587-3756
Tae Woo Kim https://orcid.org/0000-0002-2115-6448
Mo Se Kim https://orcid.org/0000-0001-7362-0030 | Fatal | ReactionOutcome | CC BY-NC | 33499678 | 19,429,456 | 2021-01 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Treatment failure'. | Hybrid laparoscopic Inokuchi shunt procedure for refractory pleural effusion and ascites associated with primary biliary cirrhosis: A case report.
BACKGROUND
Portosystemic shunts associated with portal hypertension are generally treated using non-invasive procedures, with open surgery required for refractory cases. Herein, we present a case of refractory pleural fluid and ascites associated with primary biliary cirrhosis (PBC), successfully treated using a hybrid laparoscopic Inokuchi shunt procedure.
METHODS
The patient was a 74-year-old woman with a history of PBC, presenting with breathing difficulty. Computed tomography revealed massive pleural fluid and ascites and engorged coronary and azygos veins, indicative of a portosystemic circulation shunt and a 3-cm wide hepatocellular carcinoma (HCC) on liver segment 2. The Child-Pugh score was 11. With the portosystemic shunt outflow considered as the cause of pleural fluid and ascites; therefore, we proceeded with a hybrid laparoscopic selective Inokuchi shunt procedure and tumor enucleation. Laparoscopically, the greater omentum was divided for devascularization, a 7-cm right subcostal skin incision was made, and the abdomen was opened for HCC enucleation under direct vision. The left gastric vein was divided at its junction with the portal vein and connected to the vena cava using a left external iliac vein graft through the omental foramen. After this procedure, the pleural effusion and ascites disappeared, blood ammonium level decreased to normal, and Child-Pugh score decreased to 9.
CONCLUSIONS
Using the Inokuchi shunt procedure, the portosystemic shunt, via the left gastric vein, was closed to increase portal blood flow and improve liver function.
CONCLUSIONS
As a less invasive procedure, hybrid laparoscopic approach should be considered for portosystemic shunt via the left gastric vein.
1 Introduction
Currently, first-line treatments for portosystemic shunts are non-invasive procedures, such as endoscopy or interventional radiology (IVR). However, some cases are refractory to non-invasive treatments, with impairment in liver function associated with portosystemic shunts. In these cases, surgical treatment is required. Herein, we report a case for a patient with a portosystemic shunt with massive pleural effusion and ascites associated with primary biliary cirrhosis (PBC), refractory to non-invasive treatments. A successful outcome was achieved by performing a selective Inokuchi shunt procedure [1,2]. Via a hybrid laparoscopic approach, which resulted in significant improvement in liver function, pleural effusion, and ascites. Using a hybrid laparoscopic approach is less invasive and produces superior cosmesis than an open approach.
2 Presentation of the case
Our case was prepared as per the SCARE guideline [3]. The patient was a 74-year-old Asian woman with a history of hypertension, endoscopic variceal ligation for an esophageal varicose vein, and PBC. She was referred to our hospital by a local clinic for treatment of refractory massive pleuroperitoneal fluid associated with breathing difficulty. The patient had been treated with tolvaptan, spironolactone, and furosemide. Her family history was unremarkable. A Denver shunt had been placed at the previous hospital but was already occluded. A 2-cm hepatocellular carcinoma (HCC) was observed on the surface of liver segment 2, with impaired liver function.
The patient’s body weight, height, and body mass index were 50 kg, 160 cm, and 19.5 kg/m2, respectively. On physical examination, the patient appeared alert and oriented; her abdomen was distended due to massive ascites. Her Child-Pugh score was 11, grade C, and relevant blood parameters were as follows: albumin (Alb), 2.5 g/dL; total bilirubin (T-bil), 2.0 mg/dL; prothrombin time (PT), 67 %; ammonia (NH3), 166 μg/dL; platelet count (Plt), 9.6 × 104 /μL; hemoglobin (Hb), 8.6 g/dL; and protein-induced vitamin K antagonist II (PIVKA-II), 700 mAU/mL. Chest and abdominal computed tomography revealed massive ascites and pleural fluid. The left gastric and azygos veins were engorged, indicative of a portosystemic shunt. Portal vein thrombosis and a 2-cm wide HCC, located on the surface of liver segment 2, were also identified (Fig. 1). Immediately following admission, bilateral thoracocentesis was performed to drain the pleural effusion and facilitate breathing. A hybrid laparoscopic Inokuchi shunt and enucleation of the HCC located in liver segment 2 were indicated.Fig. 1 Computed tomography images showing (a) a hepatocellular carcinoma on the surface of liver segment 2 (red arrow) and a dilated coronary vein (yellow arrow) and (b) an intravascular clot in the portal vein (blue arrow).
Fig. 1
The lead surgeon (Y.S.) had 30 years of experience. The patient was placed in the supine position, with legs spread to accommodate the assistant surgeon. Two 12-mm and three 5-mm ports were placed (Fig. 2). Under laparoscopic vision, massive ascites and a cirrhotic liver with irregular surface and nodules were observed (Fig. 3). The Denver shunt was completely occluded and, thus, removed. The greater omentum was divided along the greater curvature up to the esophagus for devascularization. Care was taken not to damage the left epiploic artery and vein. The lesser omentum was also opened, with engorgement of the left gastric vein identified (Fig. 4). A 7-cm skin incision was made in the subcostal area and the abdomen was opened. Through this window, real-time sonazoid-enhanced liver ultrasound was performed. A 2-cm HCC was detected on the surface of liver segment 2, confirming computed tomography findings; no other lesions were observed. The tumor was enucleated through this window. Subsequently, the common hepatic artery, gastroduodenal artery, proper hepatic artery, and left gastric artery were identified. The portal venous pressure, measured by direct puncture, was 30 cmH2O. To connect the vena cava and coronary vein, a 4-cm left external iliac vein graft was taken from the left inguinal area. The coronary vein was divided at the junction of the portal vein, and the orifice of the portal vein was closed using a running suture of 6-0 Proline. The left external iliac vein graft was anastomosed with the coronary vein using a running suture of 6-0 Proline in an end-to-end manner. The other end of the left external vein graft was passed through the omental foramen and anastomosed with the vena cava using a running suture of 6-0 Proline in a side-to-end manner. After the completion of anastomoses, adequate blood flow was confirmed by Doppler ultrasound. The abdomen was irrigated and hemostasis assured. Closed drainage tubes were placed in the right sub-diaphragm region and Morison’s pouch. The abdomen and left inguinal wound were closed. The patient tolerated this procedure well. The operative time was 427 min.Fig. 2 Port placements and skin incision.
Fig. 2
Fig. 3 Laparoscopic view of the cirrhotic liver showing (a) an irregular surface and nodules and (b) an engorged coronary vein.
Fig. 3
Fig. 4 Anastomosis of the external iliac vein graft with the vena cava in a side-to-end manner performed under direct vision.
Fig. 4
The postoperative course was uneventful, and the patient was discharged on postoperative day 14. One month after surgery, the pleuroperitoneal fluid disappeared and liver function improved. Her Child-Pugh score decreased to 9 and blood parameters improved as follows: NH3, 63 μg/dL; Alb, 3.0 g/dL; T-bil, 2.3 mg/dL; and PT, 73.4 %. The patient remains healthy 1 year after surgery.
3 Discussion
Our case highlights two important clinical issues. First, the Inokuchi shunt procedure increases portal blood flow and improves liver function, which led to improvement of massive pleural effusion and ascites. The hybrid laparoscopic procedure is less invasive and provides better cosmesis than an open surgical approach.
The Inokuchi shunt procedure is useful for patients with portosystemic shunt associated with portal hypertension caused by cirrhosis. The procedure shuts down the outflow from the portal vein, leading to improvement of liver function, pleural fluid, and ascites. We recognize that non-invasive procedures, such as endoscopy and IVR, are recommended as first-line treatment for esophagogastric varices shunts associated with portal hypertension. However, our patient had previously undergone esophageal varicose ligation for gastroesophageal varices. An endoscopically untreatable portosystemic shunt had developed between the left gastric and the azygous veins, leading to refractory pleural fluid and ascites associated with low liver function. The selective Inokuchi shunt procedure was used to shut down blood outflow via the coronary vein to increase portal blood flow. Compared to the distal splenorenal shunt procedure [4,5], the selective Inokuchi shunt procedure is rarely performed because of its procedural complexity. However, in this case, the main portosystemic shunt outflow was through the coronary vein, which favored the use of the selective Inokuchi shunt procedure. Importantly, we had prior experience using the selective Inokuchi shunt procedure, with good clinical outcomes achieved [6]. Splenectomy was not performed as there was no evidence of pancytopenia on blood reports. After this surgery, pleuroperitoneal fluid decreased and liver function significantly improved. The Child-Pugh score decreased from 11 to 9, downgraded to class B, and blood reports confirmed a decrease in NH3 levels to 67 μg/dL (within normal limits). Use of the Inokuchi shunt procedure also allowed us to successfully resect the HCC in liver segment 2 simultaneously.
The hybrid laparoscopic procedure is less invasive than an open surgical approach and provides better cosmesis. More importantly, our hybrid approach was more appropriate than an open approach for our patient who presented with a low nutritional status due to PBC. Among patients with portal hypertension, wound healing is slower due to edema associated with undernutrition and the risk of surgical site infection is high [7]. Therefore, total laparoscopic surgery is preferable over an open approach. However, the required anastomosis between the coronary vein and vena cava with the left iliac vein graft through omental foramen is technically demanding, requiring hand-suturing under direct vision; leakage may cause fatal bleeding and, thus, secure suture placement is paramount. The enucleation of the HCC in liver segment 2 was scheduled to be performed laparoscopically prior to surgery. However, through the 7-cm laparotomy, via the right subcostal area, and using a moving window maneuver, a clear view of the tumor was obtained and the tumor was enucleated under direct vision.
Searches performed in PubMed and Google Scholar identified no prior report of either laparoscopic or hybrid laparoscopic Inokuchi shunt procedure. To the best of our knowledge, this is the first case of this procedure reported, providing a less invasive approach for the treatment of patients with a portosystemic shunt associated with portal hypertension and refractory to non-invasive treatment. Therefore, when the coronary vein is the main outflow of the portosystemic shunt, the hybrid laparoscopic Inokuchi shunt procedure could be considered. Moreover, for patients presenting with pancytopenia, splenectomy should be performed simultaneously. When the portal blood pressure is elevated to >32 mmHg while closing the shunt, a selective shunt procedure can cause liver congestion and impair liver function [6,8].
4 Conclusion
The Inokuchi shunt procedure shuts down the portosystemic shunt to improve liver function, which leads to improvement of refractory pleural effusion and ascites associated with portal hypertension caused by liver cirrhosis. The hybrid laparoscopic approach is less invasive than the conventional open approach and provides better cosmesis. We must be aware that patients presenting with refractory pleural fluid and ascites associated with cirrhosis may have a portosystemic shunt and, thus, imaging is indicated. In our case, the hybrid laparoscopic Inokuchi shunt procedure significantly improved liver function and pleural fluid and ascites. Evaluation of long-term outcomes of this approach and studies with a greater cohort size are warranted to determine if the outcomes are equal or better than those obtained with an open surgical approach.
Declaration of Competing Interest
Nothing to declare.
Funding
Not applicable.
Ethical approval
The ethics review board of Kashiwa Kousei General Hospital approved our procedure, which conforms to the provisions of the Declaration of Helsinki.
Consent
Written informed consent was obtained from the patient for publication of this case report and any accompanying images. A copy of written consent is available for review by Editor-in-Chief of this journal on request.
Author contribution
AO and YS conceived and designed the work. KK, MK, and TS contributed to the acquisition and interpretation of the data. AO and YS drafted the paper. KK, MK, and TS revised the manuscript. All authors approved the final version of the manuscript and agree to be accountable for all aspects of the paper.
Registration of research studies
researchregistry6409.
Guarantor
Ayato Obana.
Provenance and peer review
Not commissioned and externally peer-reviewed.
Acknowledgments
I would like to thank Shigeyuki Kawa for carefully proofreading the manuscript. I am deeply grateful to Tomonori Matsumura, Shinsuke Usui, Kenichi Iwasaki, Norimasa Koide, Kazuhiro Karikomi, and Dr. Hiroaki Nomori for helpful discussions. | FUROSEMIDE, SPIRONOLACTONE, TOLVAPTAN | DrugsGivenReaction | CC BY-NC-ND | 33500229 | 19,240,496 | 2021-03 |
What is the weight of the patient? | Hybrid laparoscopic Inokuchi shunt procedure for refractory pleural effusion and ascites associated with primary biliary cirrhosis: A case report.
BACKGROUND
Portosystemic shunts associated with portal hypertension are generally treated using non-invasive procedures, with open surgery required for refractory cases. Herein, we present a case of refractory pleural fluid and ascites associated with primary biliary cirrhosis (PBC), successfully treated using a hybrid laparoscopic Inokuchi shunt procedure.
METHODS
The patient was a 74-year-old woman with a history of PBC, presenting with breathing difficulty. Computed tomography revealed massive pleural fluid and ascites and engorged coronary and azygos veins, indicative of a portosystemic circulation shunt and a 3-cm wide hepatocellular carcinoma (HCC) on liver segment 2. The Child-Pugh score was 11. With the portosystemic shunt outflow considered as the cause of pleural fluid and ascites; therefore, we proceeded with a hybrid laparoscopic selective Inokuchi shunt procedure and tumor enucleation. Laparoscopically, the greater omentum was divided for devascularization, a 7-cm right subcostal skin incision was made, and the abdomen was opened for HCC enucleation under direct vision. The left gastric vein was divided at its junction with the portal vein and connected to the vena cava using a left external iliac vein graft through the omental foramen. After this procedure, the pleural effusion and ascites disappeared, blood ammonium level decreased to normal, and Child-Pugh score decreased to 9.
CONCLUSIONS
Using the Inokuchi shunt procedure, the portosystemic shunt, via the left gastric vein, was closed to increase portal blood flow and improve liver function.
CONCLUSIONS
As a less invasive procedure, hybrid laparoscopic approach should be considered for portosystemic shunt via the left gastric vein.
1 Introduction
Currently, first-line treatments for portosystemic shunts are non-invasive procedures, such as endoscopy or interventional radiology (IVR). However, some cases are refractory to non-invasive treatments, with impairment in liver function associated with portosystemic shunts. In these cases, surgical treatment is required. Herein, we report a case for a patient with a portosystemic shunt with massive pleural effusion and ascites associated with primary biliary cirrhosis (PBC), refractory to non-invasive treatments. A successful outcome was achieved by performing a selective Inokuchi shunt procedure [1,2]. Via a hybrid laparoscopic approach, which resulted in significant improvement in liver function, pleural effusion, and ascites. Using a hybrid laparoscopic approach is less invasive and produces superior cosmesis than an open approach.
2 Presentation of the case
Our case was prepared as per the SCARE guideline [3]. The patient was a 74-year-old Asian woman with a history of hypertension, endoscopic variceal ligation for an esophageal varicose vein, and PBC. She was referred to our hospital by a local clinic for treatment of refractory massive pleuroperitoneal fluid associated with breathing difficulty. The patient had been treated with tolvaptan, spironolactone, and furosemide. Her family history was unremarkable. A Denver shunt had been placed at the previous hospital but was already occluded. A 2-cm hepatocellular carcinoma (HCC) was observed on the surface of liver segment 2, with impaired liver function.
The patient’s body weight, height, and body mass index were 50 kg, 160 cm, and 19.5 kg/m2, respectively. On physical examination, the patient appeared alert and oriented; her abdomen was distended due to massive ascites. Her Child-Pugh score was 11, grade C, and relevant blood parameters were as follows: albumin (Alb), 2.5 g/dL; total bilirubin (T-bil), 2.0 mg/dL; prothrombin time (PT), 67 %; ammonia (NH3), 166 μg/dL; platelet count (Plt), 9.6 × 104 /μL; hemoglobin (Hb), 8.6 g/dL; and protein-induced vitamin K antagonist II (PIVKA-II), 700 mAU/mL. Chest and abdominal computed tomography revealed massive ascites and pleural fluid. The left gastric and azygos veins were engorged, indicative of a portosystemic shunt. Portal vein thrombosis and a 2-cm wide HCC, located on the surface of liver segment 2, were also identified (Fig. 1). Immediately following admission, bilateral thoracocentesis was performed to drain the pleural effusion and facilitate breathing. A hybrid laparoscopic Inokuchi shunt and enucleation of the HCC located in liver segment 2 were indicated.Fig. 1 Computed tomography images showing (a) a hepatocellular carcinoma on the surface of liver segment 2 (red arrow) and a dilated coronary vein (yellow arrow) and (b) an intravascular clot in the portal vein (blue arrow).
Fig. 1
The lead surgeon (Y.S.) had 30 years of experience. The patient was placed in the supine position, with legs spread to accommodate the assistant surgeon. Two 12-mm and three 5-mm ports were placed (Fig. 2). Under laparoscopic vision, massive ascites and a cirrhotic liver with irregular surface and nodules were observed (Fig. 3). The Denver shunt was completely occluded and, thus, removed. The greater omentum was divided along the greater curvature up to the esophagus for devascularization. Care was taken not to damage the left epiploic artery and vein. The lesser omentum was also opened, with engorgement of the left gastric vein identified (Fig. 4). A 7-cm skin incision was made in the subcostal area and the abdomen was opened. Through this window, real-time sonazoid-enhanced liver ultrasound was performed. A 2-cm HCC was detected on the surface of liver segment 2, confirming computed tomography findings; no other lesions were observed. The tumor was enucleated through this window. Subsequently, the common hepatic artery, gastroduodenal artery, proper hepatic artery, and left gastric artery were identified. The portal venous pressure, measured by direct puncture, was 30 cmH2O. To connect the vena cava and coronary vein, a 4-cm left external iliac vein graft was taken from the left inguinal area. The coronary vein was divided at the junction of the portal vein, and the orifice of the portal vein was closed using a running suture of 6-0 Proline. The left external iliac vein graft was anastomosed with the coronary vein using a running suture of 6-0 Proline in an end-to-end manner. The other end of the left external vein graft was passed through the omental foramen and anastomosed with the vena cava using a running suture of 6-0 Proline in a side-to-end manner. After the completion of anastomoses, adequate blood flow was confirmed by Doppler ultrasound. The abdomen was irrigated and hemostasis assured. Closed drainage tubes were placed in the right sub-diaphragm region and Morison’s pouch. The abdomen and left inguinal wound were closed. The patient tolerated this procedure well. The operative time was 427 min.Fig. 2 Port placements and skin incision.
Fig. 2
Fig. 3 Laparoscopic view of the cirrhotic liver showing (a) an irregular surface and nodules and (b) an engorged coronary vein.
Fig. 3
Fig. 4 Anastomosis of the external iliac vein graft with the vena cava in a side-to-end manner performed under direct vision.
Fig. 4
The postoperative course was uneventful, and the patient was discharged on postoperative day 14. One month after surgery, the pleuroperitoneal fluid disappeared and liver function improved. Her Child-Pugh score decreased to 9 and blood parameters improved as follows: NH3, 63 μg/dL; Alb, 3.0 g/dL; T-bil, 2.3 mg/dL; and PT, 73.4 %. The patient remains healthy 1 year after surgery.
3 Discussion
Our case highlights two important clinical issues. First, the Inokuchi shunt procedure increases portal blood flow and improves liver function, which led to improvement of massive pleural effusion and ascites. The hybrid laparoscopic procedure is less invasive and provides better cosmesis than an open surgical approach.
The Inokuchi shunt procedure is useful for patients with portosystemic shunt associated with portal hypertension caused by cirrhosis. The procedure shuts down the outflow from the portal vein, leading to improvement of liver function, pleural fluid, and ascites. We recognize that non-invasive procedures, such as endoscopy and IVR, are recommended as first-line treatment for esophagogastric varices shunts associated with portal hypertension. However, our patient had previously undergone esophageal varicose ligation for gastroesophageal varices. An endoscopically untreatable portosystemic shunt had developed between the left gastric and the azygous veins, leading to refractory pleural fluid and ascites associated with low liver function. The selective Inokuchi shunt procedure was used to shut down blood outflow via the coronary vein to increase portal blood flow. Compared to the distal splenorenal shunt procedure [4,5], the selective Inokuchi shunt procedure is rarely performed because of its procedural complexity. However, in this case, the main portosystemic shunt outflow was through the coronary vein, which favored the use of the selective Inokuchi shunt procedure. Importantly, we had prior experience using the selective Inokuchi shunt procedure, with good clinical outcomes achieved [6]. Splenectomy was not performed as there was no evidence of pancytopenia on blood reports. After this surgery, pleuroperitoneal fluid decreased and liver function significantly improved. The Child-Pugh score decreased from 11 to 9, downgraded to class B, and blood reports confirmed a decrease in NH3 levels to 67 μg/dL (within normal limits). Use of the Inokuchi shunt procedure also allowed us to successfully resect the HCC in liver segment 2 simultaneously.
The hybrid laparoscopic procedure is less invasive than an open surgical approach and provides better cosmesis. More importantly, our hybrid approach was more appropriate than an open approach for our patient who presented with a low nutritional status due to PBC. Among patients with portal hypertension, wound healing is slower due to edema associated with undernutrition and the risk of surgical site infection is high [7]. Therefore, total laparoscopic surgery is preferable over an open approach. However, the required anastomosis between the coronary vein and vena cava with the left iliac vein graft through omental foramen is technically demanding, requiring hand-suturing under direct vision; leakage may cause fatal bleeding and, thus, secure suture placement is paramount. The enucleation of the HCC in liver segment 2 was scheduled to be performed laparoscopically prior to surgery. However, through the 7-cm laparotomy, via the right subcostal area, and using a moving window maneuver, a clear view of the tumor was obtained and the tumor was enucleated under direct vision.
Searches performed in PubMed and Google Scholar identified no prior report of either laparoscopic or hybrid laparoscopic Inokuchi shunt procedure. To the best of our knowledge, this is the first case of this procedure reported, providing a less invasive approach for the treatment of patients with a portosystemic shunt associated with portal hypertension and refractory to non-invasive treatment. Therefore, when the coronary vein is the main outflow of the portosystemic shunt, the hybrid laparoscopic Inokuchi shunt procedure could be considered. Moreover, for patients presenting with pancytopenia, splenectomy should be performed simultaneously. When the portal blood pressure is elevated to >32 mmHg while closing the shunt, a selective shunt procedure can cause liver congestion and impair liver function [6,8].
4 Conclusion
The Inokuchi shunt procedure shuts down the portosystemic shunt to improve liver function, which leads to improvement of refractory pleural effusion and ascites associated with portal hypertension caused by liver cirrhosis. The hybrid laparoscopic approach is less invasive than the conventional open approach and provides better cosmesis. We must be aware that patients presenting with refractory pleural fluid and ascites associated with cirrhosis may have a portosystemic shunt and, thus, imaging is indicated. In our case, the hybrid laparoscopic Inokuchi shunt procedure significantly improved liver function and pleural fluid and ascites. Evaluation of long-term outcomes of this approach and studies with a greater cohort size are warranted to determine if the outcomes are equal or better than those obtained with an open surgical approach.
Declaration of Competing Interest
Nothing to declare.
Funding
Not applicable.
Ethical approval
The ethics review board of Kashiwa Kousei General Hospital approved our procedure, which conforms to the provisions of the Declaration of Helsinki.
Consent
Written informed consent was obtained from the patient for publication of this case report and any accompanying images. A copy of written consent is available for review by Editor-in-Chief of this journal on request.
Author contribution
AO and YS conceived and designed the work. KK, MK, and TS contributed to the acquisition and interpretation of the data. AO and YS drafted the paper. KK, MK, and TS revised the manuscript. All authors approved the final version of the manuscript and agree to be accountable for all aspects of the paper.
Registration of research studies
researchregistry6409.
Guarantor
Ayato Obana.
Provenance and peer review
Not commissioned and externally peer-reviewed.
Acknowledgments
I would like to thank Shigeyuki Kawa for carefully proofreading the manuscript. I am deeply grateful to Tomonori Matsumura, Shinsuke Usui, Kenichi Iwasaki, Norimasa Koide, Kazuhiro Karikomi, and Dr. Hiroaki Nomori for helpful discussions. | 50 kg. | Weight | CC BY-NC-ND | 33500229 | 19,240,496 | 2021-03 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Diarrhoea'. | Gastric Cancer Harboring an ERBB3 Mutation Treated with a Pyrotinib-Irinotecan Combo: A Case Study.
Gastric cancer is common, especially in East Asian countries, and is associated with high recurrence and mortality rates. Currently, there is no standard third-line treatment for metastatic gastric cancer. In this report, we present the case of a 69-year-old man with advanced gastric cancer, whose tumor was negative for human epidermal growth factor receptor 2 (HER2) according to immunohistochemical analysis. Next-generation sequencing performed on paraffin sections of the postoperative tumor samples indicated the presence of the ERBB3 V104L mutation. The patient received irinotecan plus pyrotinib as a third-line therapy and achieved a progression-free survival of 7.6 months with a high quality of life. Therefore, the combined administration of irinotecan and pyrotinib may improve the clinical condition of patients with gastric cancer harboring an ERBB3 mutation. Moreover, ERBB3 could be a potential therapeutic target for gastric cancer.
Introduction
Gastric cancer is the second leading cause of cancer-related deaths in Asia.1 After systematic first- and second-line treatments, approximately 20–90% of patients receive active third-line or subsequent treatments;2–5 however, there are no standard advanced therapy protocols for metastatic gastric cancer, according to the National Comprehensive Cancer Network Guidelines. Preferred therapies include ramucirumab plus paclitaxel, taxane, irinotecan, TAS-102, fluorouracil plus irinotecan, apatinib, or pembrolizumab. A systematic review and meta-analysis of advanced gastric cancer indicated that the median overall survival of patients receiving third-line therapy is approximately 4.80 months compared with the 3.20 months for patients receiving only the best supportive care.6 Thus, the lack of effective third-line therapies for gastric cancer significantly restricts patient survival.
Herein, we present the case of a patient with advanced gastric cancer harboring the ERBB3 V104L mutation, who received pyrotinib plus irinotecan as a third-line therapy and achieved a progression-free survival (PFS) of 7.6 months with a high quality of life (QOL).
Case Presentation
A 69-year-old man was diagnosed with gastric adenocarcinoma in July 2015 via endoscopic biopsy. He had a family history of cancer, as his sister had colon cancer. The timeline of his treatments is shown in Figure 1. First, he underwent radical gastrectomy with postoperative pTxN1M0 grade (in another hospital). Later, from August 2015 to February 2016, the patient underwent six cycles of treatment with fluorouracil plus oxaliplatin as adjuvant chemotherapy. In October 2016, via gastroscopy, the patient was confirmed to have relapsed. Therefore, a residual gastrectomy was performed, and the postoperative stage was pT3N2M0. After the surgery, the patient received four cycles of treatment with fluorouracil plus irinotecan from December 2016 to March 2017. However, he stopped chemotherapy due to the onset of adverse events, including thrombocytopenia and diarrhea. In January 2018, he underwent positron emission tomography-computed tomography (PET-CT) due to abdominal distension. The scans showed multiple metastases in the right diaphragm and peritoneum, with a large amount of fluid in the abdominal cavity and metastasis to the liver (S5 and S6), indicating extensive disease progression. The staining results of the abdominal wall nodules are shown in Figure 2.Figure 1 The timeline of the treatment.
Abbreviations: IHC, immunohistochemistry; NGS, next-generation sequencing.Figure 2 Histologic results of abdominal wall nodules. (A) Hematoxylin-eosin staining (magnification: ×200). (B) Cytokeratin immunohistochemistry (magnification: ×200). (C) Human epidermal growth factor receptor 2 (HER2) immunohistochemistry (magnification: ×200).
In February 2018, immunohistochemical (IHC) analysis showed that the tumor was negative for human epidermal growth factor receptor 2 (HER2) (Figure 2C). The tumor tissues and matched blood samples were sent to the College of American Pathologists (CAP)-accredited and Clinical Laboratory Improvement Amendments (CLIA)-certified laboratory (OrigiMed, Shanghai, China) for targeted next-generation sequencing (NGS). Written informed consent has been obtained from the patient to have the case details and any accompanying images for publication. The genomic results revealed a mutation in ERBB3 (V104L), accompanied by mutations in TP53 (R273C), KRAS (G12F), and AMER1 (Q577*), as well as amplification of CCNE1/FOS/GATA6/MCL1/MYCN/CIC. The tumor mutational burden was 13.6 muts/Mb. Programmed death-ligand 1 (PD-L1) expression was negative and no germline mutations were detected. In March 2018, he received two courses of peritoneal thermal perfusion therapy, followed by two courses of paclitaxel peritoneal administration (240 mg). The patient was then administered six cycles of capecitabine (1 g po bid d1–14 q3w) plus oxaliplatin (230 mg ivd qd d1 q3w). In July, the CT scan suggested that the disease was stable but without significant improvement. He continued maintenance treatment with four cycles of S-1 (60 mg po bid d1–14 q3w) plus apatinib (0.25 g po qd) until October 2018. However, due to intolerable adverse events, such as incomplete small bowel obstruction, the maintenance treatment was changed to apatinib (0.5 g po qd).
In December 2018, the disease progressed with new metastases in the right adrenal gland and right paracolic sulcus revealed by CT examination. Considering that his previous NGS test indicated the presence of an ERBB3 mutation, we administered irinotecan (330 mg ivd qd d1 q2w) plus pyrotinib (320 mg qd) starting on December 28, 2018. CT scans performed on March 12 and May 9, 2019, showed stable disease (Figure 3) with decreased effusion of the abdominal cavity. The patient experienced an improvement in abdominal distension, and no additional adverse events were observed. However, on August 6, CT scans showed an increase in the abdominal metastatic tumor as well as the abdominal and pelvic effusion, which suggested disease progression. Therefore, the PFS was 7.6 months.Figure 3 The patient’s condition clinically improved after treatment with pyrotinib and irinotecan. (A) Liver S6 metastasis disappeared after treatment; (B) The diameter of the metastatic peritoneal cyst was reduced from 3.7 to 3.2 cm after treatment and maintained; (C) The diameter of the metastatic peritoneal reflex was reduced from 3.3 to 2.5 cm after treatment and maintained; (D) Ascites were significantly reduced after treatment. Red circles indicate metastatic tumor lesions.
Thereafter, the patient underwent chemotherapy, targeted therapy, and immunotherapy; however, he did not exhibit a good response. CT scans revealed multiple metastases in the abdomen and liver (S6). The patient died in March 2020.
Discussion
The benefits of the existing third-line treatments for advanced gastric cancer are limited, and many patients cannot tolerate chemotherapy-related toxicity. In recent years, targeted therapy has provided a new treatment strategy for advanced gastric cancer with more convenience and fewer side effects. However, at present, only the treatment of HER2-positive advanced gastric cancer has been effective.7 Therefore, effective therapies targeting cancer driver genes are still warranted. Herein, we report a patient with HER2-negative gastric cancer harboring an ERBB3 mutation. He received pyrotinib plus irinotecan as a third-line treatment, which resulted in a PFS of 7.6 months and a high QOL. We believe that this case provides important medical evidence for the beneficial clinical application of pan-ErbB inhibitors.
ERBB3, encoded by the ERBB3 gene, is a member of the epidermal growth factor receptor (EGFR) family. Although its intracellular tyrosine kinase domain is weak, it can still form active heterodimers with other EGFR members, thus activating pathways involved in cell proliferation and differentiation.8–10
ERBB3 mutations have been identified in some cancers, including colon and gastric cancers,11–19 which have ligand-independent and HER2-dependent transformation abilities.20 The ERBB3 V104L mutation is one of the main hotspot mutations in the extracellular domain and was identified in gallbladder cancer, rectal neuroendocrine tumors, and lung sarcomatoid carcinoma.19,21–23
Some anti-ERBB3 drugs, such as patritumab, AZD8931, and U3-1402, are still under preclinical and clinical development.24–26 Considering that ERBB3 needs to form a heterodimer with other EGFR members, antitumor drugs that target the EGFR/HER2 may be effective. Some clinical benefits have been observed with afatinib, trastuzumab plus lapatinib, and lapatinib alone, among other treatment regimens.22,27 For instance, a patient with a rectal neuroendocrine tumor harboring the ERBB3 V104L mutation was treated with trastuzumab and lapatinib as a third-line therapy, resulting in a stable disease and a PFS of 51 days.22 However, HER2-negative breast cancer patients with the ERBB3 G284R mutation, who received trastuzumab with lapatinib as a third-line treatment, showed only a partial response (PR) for more than 40 weeks.27 Additionally, a biliary tract carcinoma patient harboring an ERBB3 mutation achieved a PR after receiving trastuzumab plus lapatinib.33 Additionally, two metastatic urothelial cancers with ERBB3 V104M and G284R mutations achieved 6.3 months and 7 months of PFS, respectively, after treatment with the inhibitor afatinib (Table 1).32Table 1 Reported Cases Harboring ERBB3 Mutations Treated with Targeted Therapy
Tumor Type ERBB3 Mutation Treatment Treatment Line Response PFS Ref.
Rectal neuroendocrine tumor V104L Trastuzumab with Lapatinib Third-line SD 51 days [22]
Breast cancer G284R Trastuzumab with Lapatinib Third-line PR 40 weeks [27]
Biliary tract carcinoma – Trastuzumab with Lapatinib – PR – [32]
Metastatic urothelial cancer V104M Afatinib – SD 6.3 months [31]
Metastatic urothelial cancer G284R Afatinib – SD 7 months [31]
Abbreviations: PR, partial response; PFS, progression-free survival; SD, stable disease.
Pyrotinib is an oral, irreversible pan-ErbB inhibitor capable of blocking EGFR/HER1, HER2, and HER4 activities.28 A Phase II study showed that pyrotinib was effective in treating HER2-positive breast cancer, with a superior response to lapatinib.29 In addition, preclinical studies have confirmed that pyrotinib successfully treated non-small-cell lung carcinoma with an HER2 exon 20 mutation and HER2-positive gastric cancer.30,31,33 However, its effects on HER2-negative gastric cancer remains unknown. Here, we showed that a patient with HER2-negative gastric cancer harboring an ERBB3 mutation who received pyrotinib plus irinotecan as a third-line treatment gained a PFS of 7.6 months with a high QOL, indicating the potential of pyrotinib in treating HER2-negative gastric cancer patients with ERBB3 mutations.
One limitation of this study is that administering pyrotinib and irinotecan at the same time made it difficult to distinguish which drug produced the therapeutic effect. However, compared with the previously used fluorouracil plus irinotecan, the patient’s clinical condition was significantly improved by the irinotecan and pyrotinib combination, and his PFS reached nearly 8 months, indicating that the use of pyrotinib may have contributed to the antitumor activity by targeting the ERBB3 (V104L) mutation in this case, since pyrotinib is a pan-ErbB inhibitor. In addition, further evaluations are warranted to confirm whether pyrotinib could be widely used in gastric cancer patients with ERBB3 alterations. Collectively, we believe that ERBB3 mutations should be considered a new target for the treatment of gastric cancer.
Ethics Approval
This study was approved by the ethics committee of the Second Affiliated Hospital of Guangzhou University of Chinese Medicine. Written informed consent for this case report has been obtained from the patient.
Author Contributions
All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.
Disclosure
T.H. and W.W. declare personal fees from OrigMed outside the submitted work, and are employees of OrigiMed. The authors report no other potential conflicts of interest for this work. | FLUOROURACIL, IRINOTECAN | DrugsGivenReaction | CC BY-NC | 33500629 | 18,937,503 | 2021 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Thrombocytopenia'. | Gastric Cancer Harboring an ERBB3 Mutation Treated with a Pyrotinib-Irinotecan Combo: A Case Study.
Gastric cancer is common, especially in East Asian countries, and is associated with high recurrence and mortality rates. Currently, there is no standard third-line treatment for metastatic gastric cancer. In this report, we present the case of a 69-year-old man with advanced gastric cancer, whose tumor was negative for human epidermal growth factor receptor 2 (HER2) according to immunohistochemical analysis. Next-generation sequencing performed on paraffin sections of the postoperative tumor samples indicated the presence of the ERBB3 V104L mutation. The patient received irinotecan plus pyrotinib as a third-line therapy and achieved a progression-free survival of 7.6 months with a high quality of life. Therefore, the combined administration of irinotecan and pyrotinib may improve the clinical condition of patients with gastric cancer harboring an ERBB3 mutation. Moreover, ERBB3 could be a potential therapeutic target for gastric cancer.
Introduction
Gastric cancer is the second leading cause of cancer-related deaths in Asia.1 After systematic first- and second-line treatments, approximately 20–90% of patients receive active third-line or subsequent treatments;2–5 however, there are no standard advanced therapy protocols for metastatic gastric cancer, according to the National Comprehensive Cancer Network Guidelines. Preferred therapies include ramucirumab plus paclitaxel, taxane, irinotecan, TAS-102, fluorouracil plus irinotecan, apatinib, or pembrolizumab. A systematic review and meta-analysis of advanced gastric cancer indicated that the median overall survival of patients receiving third-line therapy is approximately 4.80 months compared with the 3.20 months for patients receiving only the best supportive care.6 Thus, the lack of effective third-line therapies for gastric cancer significantly restricts patient survival.
Herein, we present the case of a patient with advanced gastric cancer harboring the ERBB3 V104L mutation, who received pyrotinib plus irinotecan as a third-line therapy and achieved a progression-free survival (PFS) of 7.6 months with a high quality of life (QOL).
Case Presentation
A 69-year-old man was diagnosed with gastric adenocarcinoma in July 2015 via endoscopic biopsy. He had a family history of cancer, as his sister had colon cancer. The timeline of his treatments is shown in Figure 1. First, he underwent radical gastrectomy with postoperative pTxN1M0 grade (in another hospital). Later, from August 2015 to February 2016, the patient underwent six cycles of treatment with fluorouracil plus oxaliplatin as adjuvant chemotherapy. In October 2016, via gastroscopy, the patient was confirmed to have relapsed. Therefore, a residual gastrectomy was performed, and the postoperative stage was pT3N2M0. After the surgery, the patient received four cycles of treatment with fluorouracil plus irinotecan from December 2016 to March 2017. However, he stopped chemotherapy due to the onset of adverse events, including thrombocytopenia and diarrhea. In January 2018, he underwent positron emission tomography-computed tomography (PET-CT) due to abdominal distension. The scans showed multiple metastases in the right diaphragm and peritoneum, with a large amount of fluid in the abdominal cavity and metastasis to the liver (S5 and S6), indicating extensive disease progression. The staining results of the abdominal wall nodules are shown in Figure 2.Figure 1 The timeline of the treatment.
Abbreviations: IHC, immunohistochemistry; NGS, next-generation sequencing.Figure 2 Histologic results of abdominal wall nodules. (A) Hematoxylin-eosin staining (magnification: ×200). (B) Cytokeratin immunohistochemistry (magnification: ×200). (C) Human epidermal growth factor receptor 2 (HER2) immunohistochemistry (magnification: ×200).
In February 2018, immunohistochemical (IHC) analysis showed that the tumor was negative for human epidermal growth factor receptor 2 (HER2) (Figure 2C). The tumor tissues and matched blood samples were sent to the College of American Pathologists (CAP)-accredited and Clinical Laboratory Improvement Amendments (CLIA)-certified laboratory (OrigiMed, Shanghai, China) for targeted next-generation sequencing (NGS). Written informed consent has been obtained from the patient to have the case details and any accompanying images for publication. The genomic results revealed a mutation in ERBB3 (V104L), accompanied by mutations in TP53 (R273C), KRAS (G12F), and AMER1 (Q577*), as well as amplification of CCNE1/FOS/GATA6/MCL1/MYCN/CIC. The tumor mutational burden was 13.6 muts/Mb. Programmed death-ligand 1 (PD-L1) expression was negative and no germline mutations were detected. In March 2018, he received two courses of peritoneal thermal perfusion therapy, followed by two courses of paclitaxel peritoneal administration (240 mg). The patient was then administered six cycles of capecitabine (1 g po bid d1–14 q3w) plus oxaliplatin (230 mg ivd qd d1 q3w). In July, the CT scan suggested that the disease was stable but without significant improvement. He continued maintenance treatment with four cycles of S-1 (60 mg po bid d1–14 q3w) plus apatinib (0.25 g po qd) until October 2018. However, due to intolerable adverse events, such as incomplete small bowel obstruction, the maintenance treatment was changed to apatinib (0.5 g po qd).
In December 2018, the disease progressed with new metastases in the right adrenal gland and right paracolic sulcus revealed by CT examination. Considering that his previous NGS test indicated the presence of an ERBB3 mutation, we administered irinotecan (330 mg ivd qd d1 q2w) plus pyrotinib (320 mg qd) starting on December 28, 2018. CT scans performed on March 12 and May 9, 2019, showed stable disease (Figure 3) with decreased effusion of the abdominal cavity. The patient experienced an improvement in abdominal distension, and no additional adverse events were observed. However, on August 6, CT scans showed an increase in the abdominal metastatic tumor as well as the abdominal and pelvic effusion, which suggested disease progression. Therefore, the PFS was 7.6 months.Figure 3 The patient’s condition clinically improved after treatment with pyrotinib and irinotecan. (A) Liver S6 metastasis disappeared after treatment; (B) The diameter of the metastatic peritoneal cyst was reduced from 3.7 to 3.2 cm after treatment and maintained; (C) The diameter of the metastatic peritoneal reflex was reduced from 3.3 to 2.5 cm after treatment and maintained; (D) Ascites were significantly reduced after treatment. Red circles indicate metastatic tumor lesions.
Thereafter, the patient underwent chemotherapy, targeted therapy, and immunotherapy; however, he did not exhibit a good response. CT scans revealed multiple metastases in the abdomen and liver (S6). The patient died in March 2020.
Discussion
The benefits of the existing third-line treatments for advanced gastric cancer are limited, and many patients cannot tolerate chemotherapy-related toxicity. In recent years, targeted therapy has provided a new treatment strategy for advanced gastric cancer with more convenience and fewer side effects. However, at present, only the treatment of HER2-positive advanced gastric cancer has been effective.7 Therefore, effective therapies targeting cancer driver genes are still warranted. Herein, we report a patient with HER2-negative gastric cancer harboring an ERBB3 mutation. He received pyrotinib plus irinotecan as a third-line treatment, which resulted in a PFS of 7.6 months and a high QOL. We believe that this case provides important medical evidence for the beneficial clinical application of pan-ErbB inhibitors.
ERBB3, encoded by the ERBB3 gene, is a member of the epidermal growth factor receptor (EGFR) family. Although its intracellular tyrosine kinase domain is weak, it can still form active heterodimers with other EGFR members, thus activating pathways involved in cell proliferation and differentiation.8–10
ERBB3 mutations have been identified in some cancers, including colon and gastric cancers,11–19 which have ligand-independent and HER2-dependent transformation abilities.20 The ERBB3 V104L mutation is one of the main hotspot mutations in the extracellular domain and was identified in gallbladder cancer, rectal neuroendocrine tumors, and lung sarcomatoid carcinoma.19,21–23
Some anti-ERBB3 drugs, such as patritumab, AZD8931, and U3-1402, are still under preclinical and clinical development.24–26 Considering that ERBB3 needs to form a heterodimer with other EGFR members, antitumor drugs that target the EGFR/HER2 may be effective. Some clinical benefits have been observed with afatinib, trastuzumab plus lapatinib, and lapatinib alone, among other treatment regimens.22,27 For instance, a patient with a rectal neuroendocrine tumor harboring the ERBB3 V104L mutation was treated with trastuzumab and lapatinib as a third-line therapy, resulting in a stable disease and a PFS of 51 days.22 However, HER2-negative breast cancer patients with the ERBB3 G284R mutation, who received trastuzumab with lapatinib as a third-line treatment, showed only a partial response (PR) for more than 40 weeks.27 Additionally, a biliary tract carcinoma patient harboring an ERBB3 mutation achieved a PR after receiving trastuzumab plus lapatinib.33 Additionally, two metastatic urothelial cancers with ERBB3 V104M and G284R mutations achieved 6.3 months and 7 months of PFS, respectively, after treatment with the inhibitor afatinib (Table 1).32Table 1 Reported Cases Harboring ERBB3 Mutations Treated with Targeted Therapy
Tumor Type ERBB3 Mutation Treatment Treatment Line Response PFS Ref.
Rectal neuroendocrine tumor V104L Trastuzumab with Lapatinib Third-line SD 51 days [22]
Breast cancer G284R Trastuzumab with Lapatinib Third-line PR 40 weeks [27]
Biliary tract carcinoma – Trastuzumab with Lapatinib – PR – [32]
Metastatic urothelial cancer V104M Afatinib – SD 6.3 months [31]
Metastatic urothelial cancer G284R Afatinib – SD 7 months [31]
Abbreviations: PR, partial response; PFS, progression-free survival; SD, stable disease.
Pyrotinib is an oral, irreversible pan-ErbB inhibitor capable of blocking EGFR/HER1, HER2, and HER4 activities.28 A Phase II study showed that pyrotinib was effective in treating HER2-positive breast cancer, with a superior response to lapatinib.29 In addition, preclinical studies have confirmed that pyrotinib successfully treated non-small-cell lung carcinoma with an HER2 exon 20 mutation and HER2-positive gastric cancer.30,31,33 However, its effects on HER2-negative gastric cancer remains unknown. Here, we showed that a patient with HER2-negative gastric cancer harboring an ERBB3 mutation who received pyrotinib plus irinotecan as a third-line treatment gained a PFS of 7.6 months with a high QOL, indicating the potential of pyrotinib in treating HER2-negative gastric cancer patients with ERBB3 mutations.
One limitation of this study is that administering pyrotinib and irinotecan at the same time made it difficult to distinguish which drug produced the therapeutic effect. However, compared with the previously used fluorouracil plus irinotecan, the patient’s clinical condition was significantly improved by the irinotecan and pyrotinib combination, and his PFS reached nearly 8 months, indicating that the use of pyrotinib may have contributed to the antitumor activity by targeting the ERBB3 (V104L) mutation in this case, since pyrotinib is a pan-ErbB inhibitor. In addition, further evaluations are warranted to confirm whether pyrotinib could be widely used in gastric cancer patients with ERBB3 alterations. Collectively, we believe that ERBB3 mutations should be considered a new target for the treatment of gastric cancer.
Ethics Approval
This study was approved by the ethics committee of the Second Affiliated Hospital of Guangzhou University of Chinese Medicine. Written informed consent for this case report has been obtained from the patient.
Author Contributions
All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.
Disclosure
T.H. and W.W. declare personal fees from OrigMed outside the submitted work, and are employees of OrigiMed. The authors report no other potential conflicts of interest for this work. | FLUOROURACIL, IRINOTECAN | DrugsGivenReaction | CC BY-NC | 33500629 | 18,937,503 | 2021 |
What was the dosage of drug 'FLUOROURACIL'? | Gastric Cancer Harboring an ERBB3 Mutation Treated with a Pyrotinib-Irinotecan Combo: A Case Study.
Gastric cancer is common, especially in East Asian countries, and is associated with high recurrence and mortality rates. Currently, there is no standard third-line treatment for metastatic gastric cancer. In this report, we present the case of a 69-year-old man with advanced gastric cancer, whose tumor was negative for human epidermal growth factor receptor 2 (HER2) according to immunohistochemical analysis. Next-generation sequencing performed on paraffin sections of the postoperative tumor samples indicated the presence of the ERBB3 V104L mutation. The patient received irinotecan plus pyrotinib as a third-line therapy and achieved a progression-free survival of 7.6 months with a high quality of life. Therefore, the combined administration of irinotecan and pyrotinib may improve the clinical condition of patients with gastric cancer harboring an ERBB3 mutation. Moreover, ERBB3 could be a potential therapeutic target for gastric cancer.
Introduction
Gastric cancer is the second leading cause of cancer-related deaths in Asia.1 After systematic first- and second-line treatments, approximately 20–90% of patients receive active third-line or subsequent treatments;2–5 however, there are no standard advanced therapy protocols for metastatic gastric cancer, according to the National Comprehensive Cancer Network Guidelines. Preferred therapies include ramucirumab plus paclitaxel, taxane, irinotecan, TAS-102, fluorouracil plus irinotecan, apatinib, or pembrolizumab. A systematic review and meta-analysis of advanced gastric cancer indicated that the median overall survival of patients receiving third-line therapy is approximately 4.80 months compared with the 3.20 months for patients receiving only the best supportive care.6 Thus, the lack of effective third-line therapies for gastric cancer significantly restricts patient survival.
Herein, we present the case of a patient with advanced gastric cancer harboring the ERBB3 V104L mutation, who received pyrotinib plus irinotecan as a third-line therapy and achieved a progression-free survival (PFS) of 7.6 months with a high quality of life (QOL).
Case Presentation
A 69-year-old man was diagnosed with gastric adenocarcinoma in July 2015 via endoscopic biopsy. He had a family history of cancer, as his sister had colon cancer. The timeline of his treatments is shown in Figure 1. First, he underwent radical gastrectomy with postoperative pTxN1M0 grade (in another hospital). Later, from August 2015 to February 2016, the patient underwent six cycles of treatment with fluorouracil plus oxaliplatin as adjuvant chemotherapy. In October 2016, via gastroscopy, the patient was confirmed to have relapsed. Therefore, a residual gastrectomy was performed, and the postoperative stage was pT3N2M0. After the surgery, the patient received four cycles of treatment with fluorouracil plus irinotecan from December 2016 to March 2017. However, he stopped chemotherapy due to the onset of adverse events, including thrombocytopenia and diarrhea. In January 2018, he underwent positron emission tomography-computed tomography (PET-CT) due to abdominal distension. The scans showed multiple metastases in the right diaphragm and peritoneum, with a large amount of fluid in the abdominal cavity and metastasis to the liver (S5 and S6), indicating extensive disease progression. The staining results of the abdominal wall nodules are shown in Figure 2.Figure 1 The timeline of the treatment.
Abbreviations: IHC, immunohistochemistry; NGS, next-generation sequencing.Figure 2 Histologic results of abdominal wall nodules. (A) Hematoxylin-eosin staining (magnification: ×200). (B) Cytokeratin immunohistochemistry (magnification: ×200). (C) Human epidermal growth factor receptor 2 (HER2) immunohistochemistry (magnification: ×200).
In February 2018, immunohistochemical (IHC) analysis showed that the tumor was negative for human epidermal growth factor receptor 2 (HER2) (Figure 2C). The tumor tissues and matched blood samples were sent to the College of American Pathologists (CAP)-accredited and Clinical Laboratory Improvement Amendments (CLIA)-certified laboratory (OrigiMed, Shanghai, China) for targeted next-generation sequencing (NGS). Written informed consent has been obtained from the patient to have the case details and any accompanying images for publication. The genomic results revealed a mutation in ERBB3 (V104L), accompanied by mutations in TP53 (R273C), KRAS (G12F), and AMER1 (Q577*), as well as amplification of CCNE1/FOS/GATA6/MCL1/MYCN/CIC. The tumor mutational burden was 13.6 muts/Mb. Programmed death-ligand 1 (PD-L1) expression was negative and no germline mutations were detected. In March 2018, he received two courses of peritoneal thermal perfusion therapy, followed by two courses of paclitaxel peritoneal administration (240 mg). The patient was then administered six cycles of capecitabine (1 g po bid d1–14 q3w) plus oxaliplatin (230 mg ivd qd d1 q3w). In July, the CT scan suggested that the disease was stable but without significant improvement. He continued maintenance treatment with four cycles of S-1 (60 mg po bid d1–14 q3w) plus apatinib (0.25 g po qd) until October 2018. However, due to intolerable adverse events, such as incomplete small bowel obstruction, the maintenance treatment was changed to apatinib (0.5 g po qd).
In December 2018, the disease progressed with new metastases in the right adrenal gland and right paracolic sulcus revealed by CT examination. Considering that his previous NGS test indicated the presence of an ERBB3 mutation, we administered irinotecan (330 mg ivd qd d1 q2w) plus pyrotinib (320 mg qd) starting on December 28, 2018. CT scans performed on March 12 and May 9, 2019, showed stable disease (Figure 3) with decreased effusion of the abdominal cavity. The patient experienced an improvement in abdominal distension, and no additional adverse events were observed. However, on August 6, CT scans showed an increase in the abdominal metastatic tumor as well as the abdominal and pelvic effusion, which suggested disease progression. Therefore, the PFS was 7.6 months.Figure 3 The patient’s condition clinically improved after treatment with pyrotinib and irinotecan. (A) Liver S6 metastasis disappeared after treatment; (B) The diameter of the metastatic peritoneal cyst was reduced from 3.7 to 3.2 cm after treatment and maintained; (C) The diameter of the metastatic peritoneal reflex was reduced from 3.3 to 2.5 cm after treatment and maintained; (D) Ascites were significantly reduced after treatment. Red circles indicate metastatic tumor lesions.
Thereafter, the patient underwent chemotherapy, targeted therapy, and immunotherapy; however, he did not exhibit a good response. CT scans revealed multiple metastases in the abdomen and liver (S6). The patient died in March 2020.
Discussion
The benefits of the existing third-line treatments for advanced gastric cancer are limited, and many patients cannot tolerate chemotherapy-related toxicity. In recent years, targeted therapy has provided a new treatment strategy for advanced gastric cancer with more convenience and fewer side effects. However, at present, only the treatment of HER2-positive advanced gastric cancer has been effective.7 Therefore, effective therapies targeting cancer driver genes are still warranted. Herein, we report a patient with HER2-negative gastric cancer harboring an ERBB3 mutation. He received pyrotinib plus irinotecan as a third-line treatment, which resulted in a PFS of 7.6 months and a high QOL. We believe that this case provides important medical evidence for the beneficial clinical application of pan-ErbB inhibitors.
ERBB3, encoded by the ERBB3 gene, is a member of the epidermal growth factor receptor (EGFR) family. Although its intracellular tyrosine kinase domain is weak, it can still form active heterodimers with other EGFR members, thus activating pathways involved in cell proliferation and differentiation.8–10
ERBB3 mutations have been identified in some cancers, including colon and gastric cancers,11–19 which have ligand-independent and HER2-dependent transformation abilities.20 The ERBB3 V104L mutation is one of the main hotspot mutations in the extracellular domain and was identified in gallbladder cancer, rectal neuroendocrine tumors, and lung sarcomatoid carcinoma.19,21–23
Some anti-ERBB3 drugs, such as patritumab, AZD8931, and U3-1402, are still under preclinical and clinical development.24–26 Considering that ERBB3 needs to form a heterodimer with other EGFR members, antitumor drugs that target the EGFR/HER2 may be effective. Some clinical benefits have been observed with afatinib, trastuzumab plus lapatinib, and lapatinib alone, among other treatment regimens.22,27 For instance, a patient with a rectal neuroendocrine tumor harboring the ERBB3 V104L mutation was treated with trastuzumab and lapatinib as a third-line therapy, resulting in a stable disease and a PFS of 51 days.22 However, HER2-negative breast cancer patients with the ERBB3 G284R mutation, who received trastuzumab with lapatinib as a third-line treatment, showed only a partial response (PR) for more than 40 weeks.27 Additionally, a biliary tract carcinoma patient harboring an ERBB3 mutation achieved a PR after receiving trastuzumab plus lapatinib.33 Additionally, two metastatic urothelial cancers with ERBB3 V104M and G284R mutations achieved 6.3 months and 7 months of PFS, respectively, after treatment with the inhibitor afatinib (Table 1).32Table 1 Reported Cases Harboring ERBB3 Mutations Treated with Targeted Therapy
Tumor Type ERBB3 Mutation Treatment Treatment Line Response PFS Ref.
Rectal neuroendocrine tumor V104L Trastuzumab with Lapatinib Third-line SD 51 days [22]
Breast cancer G284R Trastuzumab with Lapatinib Third-line PR 40 weeks [27]
Biliary tract carcinoma – Trastuzumab with Lapatinib – PR – [32]
Metastatic urothelial cancer V104M Afatinib – SD 6.3 months [31]
Metastatic urothelial cancer G284R Afatinib – SD 7 months [31]
Abbreviations: PR, partial response; PFS, progression-free survival; SD, stable disease.
Pyrotinib is an oral, irreversible pan-ErbB inhibitor capable of blocking EGFR/HER1, HER2, and HER4 activities.28 A Phase II study showed that pyrotinib was effective in treating HER2-positive breast cancer, with a superior response to lapatinib.29 In addition, preclinical studies have confirmed that pyrotinib successfully treated non-small-cell lung carcinoma with an HER2 exon 20 mutation and HER2-positive gastric cancer.30,31,33 However, its effects on HER2-negative gastric cancer remains unknown. Here, we showed that a patient with HER2-negative gastric cancer harboring an ERBB3 mutation who received pyrotinib plus irinotecan as a third-line treatment gained a PFS of 7.6 months with a high QOL, indicating the potential of pyrotinib in treating HER2-negative gastric cancer patients with ERBB3 mutations.
One limitation of this study is that administering pyrotinib and irinotecan at the same time made it difficult to distinguish which drug produced the therapeutic effect. However, compared with the previously used fluorouracil plus irinotecan, the patient’s clinical condition was significantly improved by the irinotecan and pyrotinib combination, and his PFS reached nearly 8 months, indicating that the use of pyrotinib may have contributed to the antitumor activity by targeting the ERBB3 (V104L) mutation in this case, since pyrotinib is a pan-ErbB inhibitor. In addition, further evaluations are warranted to confirm whether pyrotinib could be widely used in gastric cancer patients with ERBB3 alterations. Collectively, we believe that ERBB3 mutations should be considered a new target for the treatment of gastric cancer.
Ethics Approval
This study was approved by the ethics committee of the Second Affiliated Hospital of Guangzhou University of Chinese Medicine. Written informed consent for this case report has been obtained from the patient.
Author Contributions
All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.
Disclosure
T.H. and W.W. declare personal fees from OrigMed outside the submitted work, and are employees of OrigiMed. The authors report no other potential conflicts of interest for this work. | DOSAGE TEXT: 4 CYCLES | DrugDosageText | CC BY-NC | 33500629 | 18,937,503 | 2021 |
What was the dosage of drug 'IRINOTECAN'? | Gastric Cancer Harboring an ERBB3 Mutation Treated with a Pyrotinib-Irinotecan Combo: A Case Study.
Gastric cancer is common, especially in East Asian countries, and is associated with high recurrence and mortality rates. Currently, there is no standard third-line treatment for metastatic gastric cancer. In this report, we present the case of a 69-year-old man with advanced gastric cancer, whose tumor was negative for human epidermal growth factor receptor 2 (HER2) according to immunohistochemical analysis. Next-generation sequencing performed on paraffin sections of the postoperative tumor samples indicated the presence of the ERBB3 V104L mutation. The patient received irinotecan plus pyrotinib as a third-line therapy and achieved a progression-free survival of 7.6 months with a high quality of life. Therefore, the combined administration of irinotecan and pyrotinib may improve the clinical condition of patients with gastric cancer harboring an ERBB3 mutation. Moreover, ERBB3 could be a potential therapeutic target for gastric cancer.
Introduction
Gastric cancer is the second leading cause of cancer-related deaths in Asia.1 After systematic first- and second-line treatments, approximately 20–90% of patients receive active third-line or subsequent treatments;2–5 however, there are no standard advanced therapy protocols for metastatic gastric cancer, according to the National Comprehensive Cancer Network Guidelines. Preferred therapies include ramucirumab plus paclitaxel, taxane, irinotecan, TAS-102, fluorouracil plus irinotecan, apatinib, or pembrolizumab. A systematic review and meta-analysis of advanced gastric cancer indicated that the median overall survival of patients receiving third-line therapy is approximately 4.80 months compared with the 3.20 months for patients receiving only the best supportive care.6 Thus, the lack of effective third-line therapies for gastric cancer significantly restricts patient survival.
Herein, we present the case of a patient with advanced gastric cancer harboring the ERBB3 V104L mutation, who received pyrotinib plus irinotecan as a third-line therapy and achieved a progression-free survival (PFS) of 7.6 months with a high quality of life (QOL).
Case Presentation
A 69-year-old man was diagnosed with gastric adenocarcinoma in July 2015 via endoscopic biopsy. He had a family history of cancer, as his sister had colon cancer. The timeline of his treatments is shown in Figure 1. First, he underwent radical gastrectomy with postoperative pTxN1M0 grade (in another hospital). Later, from August 2015 to February 2016, the patient underwent six cycles of treatment with fluorouracil plus oxaliplatin as adjuvant chemotherapy. In October 2016, via gastroscopy, the patient was confirmed to have relapsed. Therefore, a residual gastrectomy was performed, and the postoperative stage was pT3N2M0. After the surgery, the patient received four cycles of treatment with fluorouracil plus irinotecan from December 2016 to March 2017. However, he stopped chemotherapy due to the onset of adverse events, including thrombocytopenia and diarrhea. In January 2018, he underwent positron emission tomography-computed tomography (PET-CT) due to abdominal distension. The scans showed multiple metastases in the right diaphragm and peritoneum, with a large amount of fluid in the abdominal cavity and metastasis to the liver (S5 and S6), indicating extensive disease progression. The staining results of the abdominal wall nodules are shown in Figure 2.Figure 1 The timeline of the treatment.
Abbreviations: IHC, immunohistochemistry; NGS, next-generation sequencing.Figure 2 Histologic results of abdominal wall nodules. (A) Hematoxylin-eosin staining (magnification: ×200). (B) Cytokeratin immunohistochemistry (magnification: ×200). (C) Human epidermal growth factor receptor 2 (HER2) immunohistochemistry (magnification: ×200).
In February 2018, immunohistochemical (IHC) analysis showed that the tumor was negative for human epidermal growth factor receptor 2 (HER2) (Figure 2C). The tumor tissues and matched blood samples were sent to the College of American Pathologists (CAP)-accredited and Clinical Laboratory Improvement Amendments (CLIA)-certified laboratory (OrigiMed, Shanghai, China) for targeted next-generation sequencing (NGS). Written informed consent has been obtained from the patient to have the case details and any accompanying images for publication. The genomic results revealed a mutation in ERBB3 (V104L), accompanied by mutations in TP53 (R273C), KRAS (G12F), and AMER1 (Q577*), as well as amplification of CCNE1/FOS/GATA6/MCL1/MYCN/CIC. The tumor mutational burden was 13.6 muts/Mb. Programmed death-ligand 1 (PD-L1) expression was negative and no germline mutations were detected. In March 2018, he received two courses of peritoneal thermal perfusion therapy, followed by two courses of paclitaxel peritoneal administration (240 mg). The patient was then administered six cycles of capecitabine (1 g po bid d1–14 q3w) plus oxaliplatin (230 mg ivd qd d1 q3w). In July, the CT scan suggested that the disease was stable but without significant improvement. He continued maintenance treatment with four cycles of S-1 (60 mg po bid d1–14 q3w) plus apatinib (0.25 g po qd) until October 2018. However, due to intolerable adverse events, such as incomplete small bowel obstruction, the maintenance treatment was changed to apatinib (0.5 g po qd).
In December 2018, the disease progressed with new metastases in the right adrenal gland and right paracolic sulcus revealed by CT examination. Considering that his previous NGS test indicated the presence of an ERBB3 mutation, we administered irinotecan (330 mg ivd qd d1 q2w) plus pyrotinib (320 mg qd) starting on December 28, 2018. CT scans performed on March 12 and May 9, 2019, showed stable disease (Figure 3) with decreased effusion of the abdominal cavity. The patient experienced an improvement in abdominal distension, and no additional adverse events were observed. However, on August 6, CT scans showed an increase in the abdominal metastatic tumor as well as the abdominal and pelvic effusion, which suggested disease progression. Therefore, the PFS was 7.6 months.Figure 3 The patient’s condition clinically improved after treatment with pyrotinib and irinotecan. (A) Liver S6 metastasis disappeared after treatment; (B) The diameter of the metastatic peritoneal cyst was reduced from 3.7 to 3.2 cm after treatment and maintained; (C) The diameter of the metastatic peritoneal reflex was reduced from 3.3 to 2.5 cm after treatment and maintained; (D) Ascites were significantly reduced after treatment. Red circles indicate metastatic tumor lesions.
Thereafter, the patient underwent chemotherapy, targeted therapy, and immunotherapy; however, he did not exhibit a good response. CT scans revealed multiple metastases in the abdomen and liver (S6). The patient died in March 2020.
Discussion
The benefits of the existing third-line treatments for advanced gastric cancer are limited, and many patients cannot tolerate chemotherapy-related toxicity. In recent years, targeted therapy has provided a new treatment strategy for advanced gastric cancer with more convenience and fewer side effects. However, at present, only the treatment of HER2-positive advanced gastric cancer has been effective.7 Therefore, effective therapies targeting cancer driver genes are still warranted. Herein, we report a patient with HER2-negative gastric cancer harboring an ERBB3 mutation. He received pyrotinib plus irinotecan as a third-line treatment, which resulted in a PFS of 7.6 months and a high QOL. We believe that this case provides important medical evidence for the beneficial clinical application of pan-ErbB inhibitors.
ERBB3, encoded by the ERBB3 gene, is a member of the epidermal growth factor receptor (EGFR) family. Although its intracellular tyrosine kinase domain is weak, it can still form active heterodimers with other EGFR members, thus activating pathways involved in cell proliferation and differentiation.8–10
ERBB3 mutations have been identified in some cancers, including colon and gastric cancers,11–19 which have ligand-independent and HER2-dependent transformation abilities.20 The ERBB3 V104L mutation is one of the main hotspot mutations in the extracellular domain and was identified in gallbladder cancer, rectal neuroendocrine tumors, and lung sarcomatoid carcinoma.19,21–23
Some anti-ERBB3 drugs, such as patritumab, AZD8931, and U3-1402, are still under preclinical and clinical development.24–26 Considering that ERBB3 needs to form a heterodimer with other EGFR members, antitumor drugs that target the EGFR/HER2 may be effective. Some clinical benefits have been observed with afatinib, trastuzumab plus lapatinib, and lapatinib alone, among other treatment regimens.22,27 For instance, a patient with a rectal neuroendocrine tumor harboring the ERBB3 V104L mutation was treated with trastuzumab and lapatinib as a third-line therapy, resulting in a stable disease and a PFS of 51 days.22 However, HER2-negative breast cancer patients with the ERBB3 G284R mutation, who received trastuzumab with lapatinib as a third-line treatment, showed only a partial response (PR) for more than 40 weeks.27 Additionally, a biliary tract carcinoma patient harboring an ERBB3 mutation achieved a PR after receiving trastuzumab plus lapatinib.33 Additionally, two metastatic urothelial cancers with ERBB3 V104M and G284R mutations achieved 6.3 months and 7 months of PFS, respectively, after treatment with the inhibitor afatinib (Table 1).32Table 1 Reported Cases Harboring ERBB3 Mutations Treated with Targeted Therapy
Tumor Type ERBB3 Mutation Treatment Treatment Line Response PFS Ref.
Rectal neuroendocrine tumor V104L Trastuzumab with Lapatinib Third-line SD 51 days [22]
Breast cancer G284R Trastuzumab with Lapatinib Third-line PR 40 weeks [27]
Biliary tract carcinoma – Trastuzumab with Lapatinib – PR – [32]
Metastatic urothelial cancer V104M Afatinib – SD 6.3 months [31]
Metastatic urothelial cancer G284R Afatinib – SD 7 months [31]
Abbreviations: PR, partial response; PFS, progression-free survival; SD, stable disease.
Pyrotinib is an oral, irreversible pan-ErbB inhibitor capable of blocking EGFR/HER1, HER2, and HER4 activities.28 A Phase II study showed that pyrotinib was effective in treating HER2-positive breast cancer, with a superior response to lapatinib.29 In addition, preclinical studies have confirmed that pyrotinib successfully treated non-small-cell lung carcinoma with an HER2 exon 20 mutation and HER2-positive gastric cancer.30,31,33 However, its effects on HER2-negative gastric cancer remains unknown. Here, we showed that a patient with HER2-negative gastric cancer harboring an ERBB3 mutation who received pyrotinib plus irinotecan as a third-line treatment gained a PFS of 7.6 months with a high QOL, indicating the potential of pyrotinib in treating HER2-negative gastric cancer patients with ERBB3 mutations.
One limitation of this study is that administering pyrotinib and irinotecan at the same time made it difficult to distinguish which drug produced the therapeutic effect. However, compared with the previously used fluorouracil plus irinotecan, the patient’s clinical condition was significantly improved by the irinotecan and pyrotinib combination, and his PFS reached nearly 8 months, indicating that the use of pyrotinib may have contributed to the antitumor activity by targeting the ERBB3 (V104L) mutation in this case, since pyrotinib is a pan-ErbB inhibitor. In addition, further evaluations are warranted to confirm whether pyrotinib could be widely used in gastric cancer patients with ERBB3 alterations. Collectively, we believe that ERBB3 mutations should be considered a new target for the treatment of gastric cancer.
Ethics Approval
This study was approved by the ethics committee of the Second Affiliated Hospital of Guangzhou University of Chinese Medicine. Written informed consent for this case report has been obtained from the patient.
Author Contributions
All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.
Disclosure
T.H. and W.W. declare personal fees from OrigMed outside the submitted work, and are employees of OrigiMed. The authors report no other potential conflicts of interest for this work. | DOSAGE: 4 CYCLES | DrugDosageText | CC BY-NC | 33500629 | 18,937,503 | 2021 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Product use in unapproved indication'. | Chemotherapy-induced changes in tumor consistency can allow gross total resection of previously unresectable brainstem pilocytic astrocytoma.
Low-grade gliomas (LGG) are described by the World Health Organization as Grades I and II. Among LGGs, the most common primary brain tumor is pilocytic astrocytoma (PA) and carries an excellent prognosis when treated with complete surgical resection. Cases, in which this is not possible, are associated with less favorable outcomes and worse progression-free survival.
This report describes a case of a 22-year-old male, who presented with progression of a primary brainstem tumor previously treated with stereotactic radiosurgery and chemotherapy. Patient underwent surgical exploration and was diagnosed with juvenile PA, but debulking was limited by the very dense and fibrous tumor. Complete surgical resection was not possible at this time. Despite efforts to treat with chemotherapy, the patient presented a year later with clinical deterioration and severe neurologic deficits, prompting surgical re-exploration. During the second operation, the tumor was found to have undergone very significant softening in consistency, allowing for gross total resection (GTR).
Aggressive treatment of brainstem LGG should be pursued whenever possible, given its generally favorable prognosis. Repeat microsurgical resection, even with a different approach, might be reasonable and safe. Finally, chemotherapy may be associated with changes in the tumor consistency that can render previously unresectable lesions amenable to successful aggressive resection.
INTRODUCTION
Low-grade gliomas (LGG) have been described according to the World Health Organization (WHO), as Grades I and II.[7] Among these LGGs, the most common primary brain tumor is PA, accounting for 5.1% of all primary brain and central nervous system gliomas, and are described as relatively benign Grade I neoplasms.[7,9] It is the most prevalent primary brain tumor among children age 0–19 years, representing 15.3% of all cases.[9] Among children age 0–14 years, PAs have an annual age-adjusted incidence rate of 1.01 (per 100,000) that significantly decreases (0.28/100,000) in adolescents and young adults (age 15–30 years).[9] PAs in children most commonly occur in the cerebellum, optic pathway/ hypothalamic region, and supratentorial hemispheres but less frequently may occur in the brainstem and spinal cord.[2] PAs commonly arise sporadically but there is an increased risk occurrence in the optic pathway in patients with neurofibromatosis Type 1.[3]
PAs are typically well-defined tumors with cystic formation occurring within the tumor or around the tumor with associated solid nodule.[3] Histopathologically, PAs demonstrate a biphasic growth pattern with areas of highly fibrillated cells mixed with loose microcystic component accompanied by Rosenthal fibers and eosinophilic granular bodies.[3] Microvascular proliferation is common but does not indicate malignancy like in diffuse high-grade gliomas.[3] Recently, genetic alterations in the mitogen-activated protein kinase pathway of PAs have been described in literature with the most common alteration being a fusion between KIAA1549 and BRAF.[3]
While PAs are considered to have an excellent prognosis with overall 10-year survival reported over 90%, tumors where complete surgical resection is unable to be carried out have less favorable progression-free survival (PFS) and overall survival.[3,9] Primary treatment of these tumors is surgery, but radiotherapy and chemotherapy may be indicated in cases with incomplete resection or contraindication to surgery.[3] Among reported brainstem LGGs treated with surgical resection or chemotherapy, there is a clear positive correlation with extent of surgical resection and PFS [Table 1]. For example, Ahmed et al. describes 21 patients who received gross total resection (GTR) or subtotal resection (STR) with 5-year overall survival of 85%, while Klimo et al. reported 9 patients who received GTR with 5-year PFS at 75%.[1, 6]
Table 1: Summary of previously reported brainstem low-grade glioma managed with surgical resection or chemotherapy.
However, the repeat microsurgical resection of partially resected tumors that have been treated with chemotherapy resulting in gross total resection (GTR) has not been previously reported. Here, we present a case of a 22-year-old male with an initially unresectable brainstem LGG treated with vincristine and carboplatin, subsequently undergoing changes in consistency allowing for GTR on repeat surgery.
CASE PRESENTATION
History
On March 2015, a 22-year-old right-handed male with a 2 month history of right sided weakness and numbness was diagnosed with a primary brainstem tumor on MRI. He was empirically treated with temozolomide, bevacizumab, and stereotactic radiosurgery (SRS). Six months later, while stable disease was noted on brain MRI, the patient was also treated with an additional hypo-fractionated course (~50Gy in 25 fractions over 2 months) of involved field radiation.
Patient received empiric chemotherapy and radiation therapy for 21 months and follow-up brain MRI on December 2016 revealed a stable, necrotic left midbrain mass lesion measuring 2.5 × 2.1 × 1.8 cm with subacute hemorrhage in the inferior posterior margin [Figure 1].
Figure 1: Initial brain MRI: (a) Axial T1 image with contrast; (b) coronal FLAIR and (c) coronal T1 with contrast images, showing large left midbrain-pontine lesion at the time of stereotactic radiosurgery.
Nearly 2 years after the initial diagnosis, the patient presented to the clinic with progressively worsening gait instability and left greater than right weakness. Additional symptoms included transient confusion with malaise and dizziness. Brain MRI at this time revealed progression at medial and inferior margins along with interval mild ventriculomegaly consistent with obstructive hydrocephalus [Figure 2]. Decision was made to place the shunt and resect the tumor in a two-stage procedure during the same admission.
Figure 2: Two years after stereotactic radiosurgery and first round of chemotherapy: (a) Brain CT showing hemorrhage in the dorsal aspect of the tumor. Brain MRI with contrast: (b) Axial T1 image demonstrating the lesion in the left brainstem and prominent temporal horns (arrows), suggestive of an obstructive hydrocephalus; (c) sagittal T1 image showing ring-enhancing mass.
First operation – VP shunt placement and partial tumor resection/biopsy
The patient underwent an uneventful right occipital ventriculoperitoneal shunt placement with utilization of volumetric image guidance and laparoscopic assistance.
The following day, left posterior temporal-occipital craniotomy was performed with an intention to obtain tissue for diagnosis and remove as much of the brainstem tumor as safely as possible. Volumetric image guidance, microscope, neuro-monitoring, and intraoperative MRI were utilized. The brainstem lesion was approached through the posterior temporal-occipital supratentorial plane with sectioning of the tentorium to widen the exposure. The inferior posterior temporal gyrus was partially resected, allowing mobilization and preservation of the vein of Labbe. The midbrain was entered through the lateral mesencephalic safe entry zone. The tumor itself was very firm and moderately vascular. Usual microsurgical tools, including ultrasonic aspirator, were inefficient.
The tumor’s firm consistency did not allow us to remove much of the lesion. The outer portions of the mass were very fibrous and could not be mobilized. Progress was very slow. The remaining tumor was “hard as a rock” in consistency, and surgical manipulation would move it as a single block, distorting the entire brainstem. Considering the circumstances, it was felt that the safest choice would be to stop further dissection/tumor removal. Intraoperative MRI showed an approximately 40% decrease in tumor volume, but the peripherally enhancing capsule of the tumor remained unchanged, as expected given the consistency of the tumor encountered intraoperatively [Figure 3].
Figure 3: MRI images with contrast: (a) Immediate preoperative axial T1 views; (b) intraoperative Axial T1 views demonstrate an approximate 40% resection of the lesion (dotted arrow) and expected intraoperative pneumocephalus (solid arrows); (c) immediate preoperative coronal T1 views; (d) intraoperative coronal T1 views.
Pathology and postoperative course
Final pathology confirmed PA, negative for BRAF mutation or rearrangement, and the patient was started on combination therapy with carboplatin and vincristine. Unfortunately, chemotherapy was complicated with breakthrough seizures and brain MRI revealed symptomatic interval progression of the left midbrain tumor. Carboplatin and vincristine were discontinued, and seizures controlled with levetiracetam, gabapentin, and valproic acid. The patient was subsequently started on a 3 day monthly cycle of cisplatin and etoposide.
On May 2018, 14 months after the first operation, the patient presented to the emergency department for worsening headache, double vision, right-sided weakness, and gait ataxia. Clinically, he was noted to have significant right upper extremity weakness to 1–2/5, hypophonia, and rapidly progressing functional decline. Brain CT showed known mass extending from the brainstem to the thalamus, slightly eccentric on the left. A hemorrhagic component within the mass and slight increase in size of the tumor was noted as well [Figure 4].
Figure 4: Four months after partial resection and shunting; (a) non-contrast brain CT shows hemorrhagic component mostly in the posterior aspect of the tumor; (b) axial, (c) sagittal, and (d) coronal T1 MRI views with contrast, showing significant interval tumor enlargement with solid (solid arrows) and cystic (dotted arrows) components. At this point in the patient’s care, it was felt, he did not have many options left. He appeared to have either poorly tolerated, or did not respond to, chemotherapy. After two prior radiation treatments, he was ineligible for further radiation. Therefore, a multi-disciplinary decision was made to repeat neurosurgical intervention in an attempt to decompress the cyst and remove some more tumor, if feasible.
Second operation
The patient underwent suboccipital-torcular craniotomy with utilization of volumetric image guidance, microscope, and neuromonitoring. The lesion was approached through the supracerebellar infratentorial corridor through the infracollicular safe entry zone. Initially, the cystic portion was decompressed and motor oil-like fluid evacuated. The cyst walls collapsed and the brainstem relaxed. The solid portion of the tumor in the middle of the caudal midbrain and upper pons was considerably softer than in the first surgery, but more vascular with areas of hemorrhagic transformation. This tumor was successfully mobilized and grossly resected with what appeared to be normal appearing brainstem underneath.
Postoperative brain MRI demonstrated GTR and marked decompression of the brainstem. Pathology confirmed mainly necrosis and hemorrhage with focal residual PA with little/no proliferation and Ki-67 in MID1 proliferation indices [Figure 5].
Figure 5: Postoperative brain MRI: (a) Axial T1 image; (b) sagittal T1 image; and (c) coronal T1 image, demonstrating postsurgical changes after GTR of the left dorsal midbrain-pontine tumor.
Postoperative complications
Five days after surgery, the patient experienced increased somnolence. Head CT revealed new bilateral parieto-occipital subdural and epidural retrocerebellar hemorrhage. The parieto-occipital convexity hemorrhages were likely related to intra- and postoperative brain shift and tearing of the bridging veins. Although supratentorial hematomas were not felt to require surgical intervention, the retrocerebellar hemorrhage was more significant. The patient was subsequently taken up for emergent re-opening of the torcular-suboccipital craniotomy for evacuation of the epidural hematoma. Postoperative CT revealed successful evacuation of the extra-axial hematoma.
Three months after surgery, follow-up brain MRI demonstrated evolving postoperative changes with no residual/recurrent tumor.
Twelve months after surgery, the patient’s dysarthria improved. Motor exam revealed persistent right hemiparesis (improved to 3–4/5), arm weaker than the leg. Patient was now able to stand and walk short distances with assistance. Chronic steroids have been successfully weaned off, and the patient had a complete reversal of cushingoid appearance.
DISCUSSION
LGG have been described according to the WHO, as Grades I and II.[7] Among these LGGs, the most common primary brain tumor is PA described as a relatively benign Grade I neoplasm.[7] The most common clinical presentation of brainstem PA is a focal neurological deficit of cranial nerves corresponding to the anatomical location of the tumor with or without involvement of the motor/sensory long tracts.[2] Additional symptoms of hydrocephalus and headache may be present as well.[2] While PAs are considered to have an excellent prognosis with overall 10-year survival reported over 90%, tumors where complete surgical resection is unable to be carried out have less favorable PFS and overall survival.[3,9] In addition, brainstem PA is complicated with significant neurological morbidity secondary to their location.[11]
The gold standard for approaching pediatric brainstem LGGs is initial treatment with safe maximal surgical resection, in conjunction with chemotherapy and radiation therapy as needed.[11] Upadhyaya et al. studied 25 cases of pediatric non tectal brainstem LGGs treated with surgical resection. All three patients with GTR and histological diagnosis of PA demonstrated no evidence of disease at long-term follow-up.[11] Overall, the study found a 10-year PFS rate of 71% in these 25 patients, of which 16 were cases of brainstem PA.[11]
Kestle et al.’s study investigated 28 cases of pediatric brainstem PA, of which 25 were treated with neurosurgical resection.[5] Kestle et al.’s study found a direct correlation between extent of tumor resection and long-term survival.[5] Twelve patients had postoperative imaging demonstrating GTR or linear enhancement, with a 10-year PFS of 62%. In contrast, the 13 patients with postoperative scans revealing residual solid tumor only had a 10-year PFE of 19%.[5] While patients with GTR saw long-term PFS, surgical resection was often associated with neurological dysfunction most apparent in the immediate postoperative phase.[5] This deficit typically resolved but persisted in 6 of the 28 patients.[5]
While GTR of brainstem PA clearly leads to favorable outcomes, adjuvant or neoadjuvant radiation therapy may also be used in the management of these tumors. Gagliardi et al. investigated the use of radiotherapy for LGGs in 39 patients.[4] Of these, four solid lesions converted into mixed tumors following GKRS.[4] In addition, tumor progression was followed in patients and found cyst progression occurred more commonly in patients with a mixed lesion at the time of radiation therapy.[4] Four of five reported incidents of cyst progression were found to be PA.[4] Overall, this study found a 10-year PFS rate of 39.1% in those treated with Gamma Knife Radiosurgery.[4]
The patient in this case report was not offered surgery as the first line of treatment. Instead, he was treated with SRS, another hypofractionated course of radiotherapy, and chemotherapy. It is conceivable that the tumor could acquire its very firm consistency as a consequence of these treatments. Hardening of the tumor due to radiation therapy is a well described phenomenon.[10] Multiple authors reported surgical challenges associated with resection of previously irradiated tumors. Changes in the tumor consistency might be related to the development of fibrosis within the tumor and excessive perifocal scarring. Such changes were noted during the first surgery. However, the second round of chemotherapy with carboplatin and vincristine, along with another cycle with cisplatin and etoposide, was associated with considerable softening of the tumor and its partial hemorrhagic transformation that allowed for gross resection. This phenomenon has not been previously described in the literature. It could be also suggested that between [Figures 3 and 4], a newly enhancing tissue (best seen on the axial and coronal MR images) may represent a growth of radiotherapynaïve tumor, which was softer and thus, more amenable to surgical resection. This hypothesis can be true only if the surgeons did not leave behind a considerable amount of solid tissue after the first surgical procedure, which was not the case. In addition, the tumor, new or old, could outgrow its blood supply and undergo necrotic transformation, which could also be influenced by chemotherapy. It is important to note, though, that the new radiotherapy-naïve component was exposed to chemotherapy, much like the hard tumor treated with XRT. At this point, we can only speculate about the extent to which the above potential mechanisms could have contributed to the apparent softening of the mass. Although not unequivocally proving the exact causality, the facts that remain clear are (1) very hard consistency of the tumor discovered on the first surgery and (2) a quite significant softening of previously unresectable lesion after chemotherapy. Therefore, the authors believe that chemotherapy could have contributed to the success of the second surgery.
We did not find any reports of repeat surgical interventions on adult patients with brainstem LGG. One study reported 12 pediatric patients with midbrain LGG who were treated with multiple repeat tumor resections due to disease progression with seemingly stable long-term results obtained in at least nine patients.[8] The surgeons treating brainstem LGG should be aware of possibility of chemotherapy induced tumor softening phenomenon and be prepared to offer repeat surgery, even if the lesion was initially deemed unresectable.
CONCLUSION
Aggressive treatment of brainstem LGG should be pursued whenever possible, given its generally favorable prognosis. Repeat microsurgical resection, even with a different approach, might be reasonable and safe. Finally, chemotherapy may be associated with changes in the tumor consistency that can render previously unresectable lesions amenable to successful aggressive resection.
How to cite this article: Chung DJ, Arif B, Odia Y, Siomin V. Chemotherapy-induced changes in tumor consistency can allow gross total resection of previously unresectable brainstem pilocytic astrocytoma. Surg Neurol Int 2021;12:12.
Declaration of patient consent
The authors certify that they have obtained all appropriate patient consent.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest. | BEVACIZUMAB, TEMOZOLOMIDE | DrugsGivenReaction | CC BY-NC-SA | 33500827 | 19,408,198 | 2021 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Seizure'. | Chemotherapy-induced changes in tumor consistency can allow gross total resection of previously unresectable brainstem pilocytic astrocytoma.
Low-grade gliomas (LGG) are described by the World Health Organization as Grades I and II. Among LGGs, the most common primary brain tumor is pilocytic astrocytoma (PA) and carries an excellent prognosis when treated with complete surgical resection. Cases, in which this is not possible, are associated with less favorable outcomes and worse progression-free survival.
This report describes a case of a 22-year-old male, who presented with progression of a primary brainstem tumor previously treated with stereotactic radiosurgery and chemotherapy. Patient underwent surgical exploration and was diagnosed with juvenile PA, but debulking was limited by the very dense and fibrous tumor. Complete surgical resection was not possible at this time. Despite efforts to treat with chemotherapy, the patient presented a year later with clinical deterioration and severe neurologic deficits, prompting surgical re-exploration. During the second operation, the tumor was found to have undergone very significant softening in consistency, allowing for gross total resection (GTR).
Aggressive treatment of brainstem LGG should be pursued whenever possible, given its generally favorable prognosis. Repeat microsurgical resection, even with a different approach, might be reasonable and safe. Finally, chemotherapy may be associated with changes in the tumor consistency that can render previously unresectable lesions amenable to successful aggressive resection.
INTRODUCTION
Low-grade gliomas (LGG) have been described according to the World Health Organization (WHO), as Grades I and II.[7] Among these LGGs, the most common primary brain tumor is PA, accounting for 5.1% of all primary brain and central nervous system gliomas, and are described as relatively benign Grade I neoplasms.[7,9] It is the most prevalent primary brain tumor among children age 0–19 years, representing 15.3% of all cases.[9] Among children age 0–14 years, PAs have an annual age-adjusted incidence rate of 1.01 (per 100,000) that significantly decreases (0.28/100,000) in adolescents and young adults (age 15–30 years).[9] PAs in children most commonly occur in the cerebellum, optic pathway/ hypothalamic region, and supratentorial hemispheres but less frequently may occur in the brainstem and spinal cord.[2] PAs commonly arise sporadically but there is an increased risk occurrence in the optic pathway in patients with neurofibromatosis Type 1.[3]
PAs are typically well-defined tumors with cystic formation occurring within the tumor or around the tumor with associated solid nodule.[3] Histopathologically, PAs demonstrate a biphasic growth pattern with areas of highly fibrillated cells mixed with loose microcystic component accompanied by Rosenthal fibers and eosinophilic granular bodies.[3] Microvascular proliferation is common but does not indicate malignancy like in diffuse high-grade gliomas.[3] Recently, genetic alterations in the mitogen-activated protein kinase pathway of PAs have been described in literature with the most common alteration being a fusion between KIAA1549 and BRAF.[3]
While PAs are considered to have an excellent prognosis with overall 10-year survival reported over 90%, tumors where complete surgical resection is unable to be carried out have less favorable progression-free survival (PFS) and overall survival.[3,9] Primary treatment of these tumors is surgery, but radiotherapy and chemotherapy may be indicated in cases with incomplete resection or contraindication to surgery.[3] Among reported brainstem LGGs treated with surgical resection or chemotherapy, there is a clear positive correlation with extent of surgical resection and PFS [Table 1]. For example, Ahmed et al. describes 21 patients who received gross total resection (GTR) or subtotal resection (STR) with 5-year overall survival of 85%, while Klimo et al. reported 9 patients who received GTR with 5-year PFS at 75%.[1, 6]
Table 1: Summary of previously reported brainstem low-grade glioma managed with surgical resection or chemotherapy.
However, the repeat microsurgical resection of partially resected tumors that have been treated with chemotherapy resulting in gross total resection (GTR) has not been previously reported. Here, we present a case of a 22-year-old male with an initially unresectable brainstem LGG treated with vincristine and carboplatin, subsequently undergoing changes in consistency allowing for GTR on repeat surgery.
CASE PRESENTATION
History
On March 2015, a 22-year-old right-handed male with a 2 month history of right sided weakness and numbness was diagnosed with a primary brainstem tumor on MRI. He was empirically treated with temozolomide, bevacizumab, and stereotactic radiosurgery (SRS). Six months later, while stable disease was noted on brain MRI, the patient was also treated with an additional hypo-fractionated course (~50Gy in 25 fractions over 2 months) of involved field radiation.
Patient received empiric chemotherapy and radiation therapy for 21 months and follow-up brain MRI on December 2016 revealed a stable, necrotic left midbrain mass lesion measuring 2.5 × 2.1 × 1.8 cm with subacute hemorrhage in the inferior posterior margin [Figure 1].
Figure 1: Initial brain MRI: (a) Axial T1 image with contrast; (b) coronal FLAIR and (c) coronal T1 with contrast images, showing large left midbrain-pontine lesion at the time of stereotactic radiosurgery.
Nearly 2 years after the initial diagnosis, the patient presented to the clinic with progressively worsening gait instability and left greater than right weakness. Additional symptoms included transient confusion with malaise and dizziness. Brain MRI at this time revealed progression at medial and inferior margins along with interval mild ventriculomegaly consistent with obstructive hydrocephalus [Figure 2]. Decision was made to place the shunt and resect the tumor in a two-stage procedure during the same admission.
Figure 2: Two years after stereotactic radiosurgery and first round of chemotherapy: (a) Brain CT showing hemorrhage in the dorsal aspect of the tumor. Brain MRI with contrast: (b) Axial T1 image demonstrating the lesion in the left brainstem and prominent temporal horns (arrows), suggestive of an obstructive hydrocephalus; (c) sagittal T1 image showing ring-enhancing mass.
First operation – VP shunt placement and partial tumor resection/biopsy
The patient underwent an uneventful right occipital ventriculoperitoneal shunt placement with utilization of volumetric image guidance and laparoscopic assistance.
The following day, left posterior temporal-occipital craniotomy was performed with an intention to obtain tissue for diagnosis and remove as much of the brainstem tumor as safely as possible. Volumetric image guidance, microscope, neuro-monitoring, and intraoperative MRI were utilized. The brainstem lesion was approached through the posterior temporal-occipital supratentorial plane with sectioning of the tentorium to widen the exposure. The inferior posterior temporal gyrus was partially resected, allowing mobilization and preservation of the vein of Labbe. The midbrain was entered through the lateral mesencephalic safe entry zone. The tumor itself was very firm and moderately vascular. Usual microsurgical tools, including ultrasonic aspirator, were inefficient.
The tumor’s firm consistency did not allow us to remove much of the lesion. The outer portions of the mass were very fibrous and could not be mobilized. Progress was very slow. The remaining tumor was “hard as a rock” in consistency, and surgical manipulation would move it as a single block, distorting the entire brainstem. Considering the circumstances, it was felt that the safest choice would be to stop further dissection/tumor removal. Intraoperative MRI showed an approximately 40% decrease in tumor volume, but the peripherally enhancing capsule of the tumor remained unchanged, as expected given the consistency of the tumor encountered intraoperatively [Figure 3].
Figure 3: MRI images with contrast: (a) Immediate preoperative axial T1 views; (b) intraoperative Axial T1 views demonstrate an approximate 40% resection of the lesion (dotted arrow) and expected intraoperative pneumocephalus (solid arrows); (c) immediate preoperative coronal T1 views; (d) intraoperative coronal T1 views.
Pathology and postoperative course
Final pathology confirmed PA, negative for BRAF mutation or rearrangement, and the patient was started on combination therapy with carboplatin and vincristine. Unfortunately, chemotherapy was complicated with breakthrough seizures and brain MRI revealed symptomatic interval progression of the left midbrain tumor. Carboplatin and vincristine were discontinued, and seizures controlled with levetiracetam, gabapentin, and valproic acid. The patient was subsequently started on a 3 day monthly cycle of cisplatin and etoposide.
On May 2018, 14 months after the first operation, the patient presented to the emergency department for worsening headache, double vision, right-sided weakness, and gait ataxia. Clinically, he was noted to have significant right upper extremity weakness to 1–2/5, hypophonia, and rapidly progressing functional decline. Brain CT showed known mass extending from the brainstem to the thalamus, slightly eccentric on the left. A hemorrhagic component within the mass and slight increase in size of the tumor was noted as well [Figure 4].
Figure 4: Four months after partial resection and shunting; (a) non-contrast brain CT shows hemorrhagic component mostly in the posterior aspect of the tumor; (b) axial, (c) sagittal, and (d) coronal T1 MRI views with contrast, showing significant interval tumor enlargement with solid (solid arrows) and cystic (dotted arrows) components. At this point in the patient’s care, it was felt, he did not have many options left. He appeared to have either poorly tolerated, or did not respond to, chemotherapy. After two prior radiation treatments, he was ineligible for further radiation. Therefore, a multi-disciplinary decision was made to repeat neurosurgical intervention in an attempt to decompress the cyst and remove some more tumor, if feasible.
Second operation
The patient underwent suboccipital-torcular craniotomy with utilization of volumetric image guidance, microscope, and neuromonitoring. The lesion was approached through the supracerebellar infratentorial corridor through the infracollicular safe entry zone. Initially, the cystic portion was decompressed and motor oil-like fluid evacuated. The cyst walls collapsed and the brainstem relaxed. The solid portion of the tumor in the middle of the caudal midbrain and upper pons was considerably softer than in the first surgery, but more vascular with areas of hemorrhagic transformation. This tumor was successfully mobilized and grossly resected with what appeared to be normal appearing brainstem underneath.
Postoperative brain MRI demonstrated GTR and marked decompression of the brainstem. Pathology confirmed mainly necrosis and hemorrhage with focal residual PA with little/no proliferation and Ki-67 in MID1 proliferation indices [Figure 5].
Figure 5: Postoperative brain MRI: (a) Axial T1 image; (b) sagittal T1 image; and (c) coronal T1 image, demonstrating postsurgical changes after GTR of the left dorsal midbrain-pontine tumor.
Postoperative complications
Five days after surgery, the patient experienced increased somnolence. Head CT revealed new bilateral parieto-occipital subdural and epidural retrocerebellar hemorrhage. The parieto-occipital convexity hemorrhages were likely related to intra- and postoperative brain shift and tearing of the bridging veins. Although supratentorial hematomas were not felt to require surgical intervention, the retrocerebellar hemorrhage was more significant. The patient was subsequently taken up for emergent re-opening of the torcular-suboccipital craniotomy for evacuation of the epidural hematoma. Postoperative CT revealed successful evacuation of the extra-axial hematoma.
Three months after surgery, follow-up brain MRI demonstrated evolving postoperative changes with no residual/recurrent tumor.
Twelve months after surgery, the patient’s dysarthria improved. Motor exam revealed persistent right hemiparesis (improved to 3–4/5), arm weaker than the leg. Patient was now able to stand and walk short distances with assistance. Chronic steroids have been successfully weaned off, and the patient had a complete reversal of cushingoid appearance.
DISCUSSION
LGG have been described according to the WHO, as Grades I and II.[7] Among these LGGs, the most common primary brain tumor is PA described as a relatively benign Grade I neoplasm.[7] The most common clinical presentation of brainstem PA is a focal neurological deficit of cranial nerves corresponding to the anatomical location of the tumor with or without involvement of the motor/sensory long tracts.[2] Additional symptoms of hydrocephalus and headache may be present as well.[2] While PAs are considered to have an excellent prognosis with overall 10-year survival reported over 90%, tumors where complete surgical resection is unable to be carried out have less favorable PFS and overall survival.[3,9] In addition, brainstem PA is complicated with significant neurological morbidity secondary to their location.[11]
The gold standard for approaching pediatric brainstem LGGs is initial treatment with safe maximal surgical resection, in conjunction with chemotherapy and radiation therapy as needed.[11] Upadhyaya et al. studied 25 cases of pediatric non tectal brainstem LGGs treated with surgical resection. All three patients with GTR and histological diagnosis of PA demonstrated no evidence of disease at long-term follow-up.[11] Overall, the study found a 10-year PFS rate of 71% in these 25 patients, of which 16 were cases of brainstem PA.[11]
Kestle et al.’s study investigated 28 cases of pediatric brainstem PA, of which 25 were treated with neurosurgical resection.[5] Kestle et al.’s study found a direct correlation between extent of tumor resection and long-term survival.[5] Twelve patients had postoperative imaging demonstrating GTR or linear enhancement, with a 10-year PFS of 62%. In contrast, the 13 patients with postoperative scans revealing residual solid tumor only had a 10-year PFE of 19%.[5] While patients with GTR saw long-term PFS, surgical resection was often associated with neurological dysfunction most apparent in the immediate postoperative phase.[5] This deficit typically resolved but persisted in 6 of the 28 patients.[5]
While GTR of brainstem PA clearly leads to favorable outcomes, adjuvant or neoadjuvant radiation therapy may also be used in the management of these tumors. Gagliardi et al. investigated the use of radiotherapy for LGGs in 39 patients.[4] Of these, four solid lesions converted into mixed tumors following GKRS.[4] In addition, tumor progression was followed in patients and found cyst progression occurred more commonly in patients with a mixed lesion at the time of radiation therapy.[4] Four of five reported incidents of cyst progression were found to be PA.[4] Overall, this study found a 10-year PFS rate of 39.1% in those treated with Gamma Knife Radiosurgery.[4]
The patient in this case report was not offered surgery as the first line of treatment. Instead, he was treated with SRS, another hypofractionated course of radiotherapy, and chemotherapy. It is conceivable that the tumor could acquire its very firm consistency as a consequence of these treatments. Hardening of the tumor due to radiation therapy is a well described phenomenon.[10] Multiple authors reported surgical challenges associated with resection of previously irradiated tumors. Changes in the tumor consistency might be related to the development of fibrosis within the tumor and excessive perifocal scarring. Such changes were noted during the first surgery. However, the second round of chemotherapy with carboplatin and vincristine, along with another cycle with cisplatin and etoposide, was associated with considerable softening of the tumor and its partial hemorrhagic transformation that allowed for gross resection. This phenomenon has not been previously described in the literature. It could be also suggested that between [Figures 3 and 4], a newly enhancing tissue (best seen on the axial and coronal MR images) may represent a growth of radiotherapynaïve tumor, which was softer and thus, more amenable to surgical resection. This hypothesis can be true only if the surgeons did not leave behind a considerable amount of solid tissue after the first surgical procedure, which was not the case. In addition, the tumor, new or old, could outgrow its blood supply and undergo necrotic transformation, which could also be influenced by chemotherapy. It is important to note, though, that the new radiotherapy-naïve component was exposed to chemotherapy, much like the hard tumor treated with XRT. At this point, we can only speculate about the extent to which the above potential mechanisms could have contributed to the apparent softening of the mass. Although not unequivocally proving the exact causality, the facts that remain clear are (1) very hard consistency of the tumor discovered on the first surgery and (2) a quite significant softening of previously unresectable lesion after chemotherapy. Therefore, the authors believe that chemotherapy could have contributed to the success of the second surgery.
We did not find any reports of repeat surgical interventions on adult patients with brainstem LGG. One study reported 12 pediatric patients with midbrain LGG who were treated with multiple repeat tumor resections due to disease progression with seemingly stable long-term results obtained in at least nine patients.[8] The surgeons treating brainstem LGG should be aware of possibility of chemotherapy induced tumor softening phenomenon and be prepared to offer repeat surgery, even if the lesion was initially deemed unresectable.
CONCLUSION
Aggressive treatment of brainstem LGG should be pursued whenever possible, given its generally favorable prognosis. Repeat microsurgical resection, even with a different approach, might be reasonable and safe. Finally, chemotherapy may be associated with changes in the tumor consistency that can render previously unresectable lesions amenable to successful aggressive resection.
How to cite this article: Chung DJ, Arif B, Odia Y, Siomin V. Chemotherapy-induced changes in tumor consistency can allow gross total resection of previously unresectable brainstem pilocytic astrocytoma. Surg Neurol Int 2021;12:12.
Declaration of patient consent
The authors certify that they have obtained all appropriate patient consent.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest. | CARBOPLATIN, VINCRISTINE | DrugsGivenReaction | CC BY-NC-SA | 33500827 | 19,382,011 | 2021 |
What was the administration route of drug 'TEMOZOLOMIDE'? | Chemotherapy-induced changes in tumor consistency can allow gross total resection of previously unresectable brainstem pilocytic astrocytoma.
Low-grade gliomas (LGG) are described by the World Health Organization as Grades I and II. Among LGGs, the most common primary brain tumor is pilocytic astrocytoma (PA) and carries an excellent prognosis when treated with complete surgical resection. Cases, in which this is not possible, are associated with less favorable outcomes and worse progression-free survival.
This report describes a case of a 22-year-old male, who presented with progression of a primary brainstem tumor previously treated with stereotactic radiosurgery and chemotherapy. Patient underwent surgical exploration and was diagnosed with juvenile PA, but debulking was limited by the very dense and fibrous tumor. Complete surgical resection was not possible at this time. Despite efforts to treat with chemotherapy, the patient presented a year later with clinical deterioration and severe neurologic deficits, prompting surgical re-exploration. During the second operation, the tumor was found to have undergone very significant softening in consistency, allowing for gross total resection (GTR).
Aggressive treatment of brainstem LGG should be pursued whenever possible, given its generally favorable prognosis. Repeat microsurgical resection, even with a different approach, might be reasonable and safe. Finally, chemotherapy may be associated with changes in the tumor consistency that can render previously unresectable lesions amenable to successful aggressive resection.
INTRODUCTION
Low-grade gliomas (LGG) have been described according to the World Health Organization (WHO), as Grades I and II.[7] Among these LGGs, the most common primary brain tumor is PA, accounting for 5.1% of all primary brain and central nervous system gliomas, and are described as relatively benign Grade I neoplasms.[7,9] It is the most prevalent primary brain tumor among children age 0–19 years, representing 15.3% of all cases.[9] Among children age 0–14 years, PAs have an annual age-adjusted incidence rate of 1.01 (per 100,000) that significantly decreases (0.28/100,000) in adolescents and young adults (age 15–30 years).[9] PAs in children most commonly occur in the cerebellum, optic pathway/ hypothalamic region, and supratentorial hemispheres but less frequently may occur in the brainstem and spinal cord.[2] PAs commonly arise sporadically but there is an increased risk occurrence in the optic pathway in patients with neurofibromatosis Type 1.[3]
PAs are typically well-defined tumors with cystic formation occurring within the tumor or around the tumor with associated solid nodule.[3] Histopathologically, PAs demonstrate a biphasic growth pattern with areas of highly fibrillated cells mixed with loose microcystic component accompanied by Rosenthal fibers and eosinophilic granular bodies.[3] Microvascular proliferation is common but does not indicate malignancy like in diffuse high-grade gliomas.[3] Recently, genetic alterations in the mitogen-activated protein kinase pathway of PAs have been described in literature with the most common alteration being a fusion between KIAA1549 and BRAF.[3]
While PAs are considered to have an excellent prognosis with overall 10-year survival reported over 90%, tumors where complete surgical resection is unable to be carried out have less favorable progression-free survival (PFS) and overall survival.[3,9] Primary treatment of these tumors is surgery, but radiotherapy and chemotherapy may be indicated in cases with incomplete resection or contraindication to surgery.[3] Among reported brainstem LGGs treated with surgical resection or chemotherapy, there is a clear positive correlation with extent of surgical resection and PFS [Table 1]. For example, Ahmed et al. describes 21 patients who received gross total resection (GTR) or subtotal resection (STR) with 5-year overall survival of 85%, while Klimo et al. reported 9 patients who received GTR with 5-year PFS at 75%.[1, 6]
Table 1: Summary of previously reported brainstem low-grade glioma managed with surgical resection or chemotherapy.
However, the repeat microsurgical resection of partially resected tumors that have been treated with chemotherapy resulting in gross total resection (GTR) has not been previously reported. Here, we present a case of a 22-year-old male with an initially unresectable brainstem LGG treated with vincristine and carboplatin, subsequently undergoing changes in consistency allowing for GTR on repeat surgery.
CASE PRESENTATION
History
On March 2015, a 22-year-old right-handed male with a 2 month history of right sided weakness and numbness was diagnosed with a primary brainstem tumor on MRI. He was empirically treated with temozolomide, bevacizumab, and stereotactic radiosurgery (SRS). Six months later, while stable disease was noted on brain MRI, the patient was also treated with an additional hypo-fractionated course (~50Gy in 25 fractions over 2 months) of involved field radiation.
Patient received empiric chemotherapy and radiation therapy for 21 months and follow-up brain MRI on December 2016 revealed a stable, necrotic left midbrain mass lesion measuring 2.5 × 2.1 × 1.8 cm with subacute hemorrhage in the inferior posterior margin [Figure 1].
Figure 1: Initial brain MRI: (a) Axial T1 image with contrast; (b) coronal FLAIR and (c) coronal T1 with contrast images, showing large left midbrain-pontine lesion at the time of stereotactic radiosurgery.
Nearly 2 years after the initial diagnosis, the patient presented to the clinic with progressively worsening gait instability and left greater than right weakness. Additional symptoms included transient confusion with malaise and dizziness. Brain MRI at this time revealed progression at medial and inferior margins along with interval mild ventriculomegaly consistent with obstructive hydrocephalus [Figure 2]. Decision was made to place the shunt and resect the tumor in a two-stage procedure during the same admission.
Figure 2: Two years after stereotactic radiosurgery and first round of chemotherapy: (a) Brain CT showing hemorrhage in the dorsal aspect of the tumor. Brain MRI with contrast: (b) Axial T1 image demonstrating the lesion in the left brainstem and prominent temporal horns (arrows), suggestive of an obstructive hydrocephalus; (c) sagittal T1 image showing ring-enhancing mass.
First operation – VP shunt placement and partial tumor resection/biopsy
The patient underwent an uneventful right occipital ventriculoperitoneal shunt placement with utilization of volumetric image guidance and laparoscopic assistance.
The following day, left posterior temporal-occipital craniotomy was performed with an intention to obtain tissue for diagnosis and remove as much of the brainstem tumor as safely as possible. Volumetric image guidance, microscope, neuro-monitoring, and intraoperative MRI were utilized. The brainstem lesion was approached through the posterior temporal-occipital supratentorial plane with sectioning of the tentorium to widen the exposure. The inferior posterior temporal gyrus was partially resected, allowing mobilization and preservation of the vein of Labbe. The midbrain was entered through the lateral mesencephalic safe entry zone. The tumor itself was very firm and moderately vascular. Usual microsurgical tools, including ultrasonic aspirator, were inefficient.
The tumor’s firm consistency did not allow us to remove much of the lesion. The outer portions of the mass were very fibrous and could not be mobilized. Progress was very slow. The remaining tumor was “hard as a rock” in consistency, and surgical manipulation would move it as a single block, distorting the entire brainstem. Considering the circumstances, it was felt that the safest choice would be to stop further dissection/tumor removal. Intraoperative MRI showed an approximately 40% decrease in tumor volume, but the peripherally enhancing capsule of the tumor remained unchanged, as expected given the consistency of the tumor encountered intraoperatively [Figure 3].
Figure 3: MRI images with contrast: (a) Immediate preoperative axial T1 views; (b) intraoperative Axial T1 views demonstrate an approximate 40% resection of the lesion (dotted arrow) and expected intraoperative pneumocephalus (solid arrows); (c) immediate preoperative coronal T1 views; (d) intraoperative coronal T1 views.
Pathology and postoperative course
Final pathology confirmed PA, negative for BRAF mutation or rearrangement, and the patient was started on combination therapy with carboplatin and vincristine. Unfortunately, chemotherapy was complicated with breakthrough seizures and brain MRI revealed symptomatic interval progression of the left midbrain tumor. Carboplatin and vincristine were discontinued, and seizures controlled with levetiracetam, gabapentin, and valproic acid. The patient was subsequently started on a 3 day monthly cycle of cisplatin and etoposide.
On May 2018, 14 months after the first operation, the patient presented to the emergency department for worsening headache, double vision, right-sided weakness, and gait ataxia. Clinically, he was noted to have significant right upper extremity weakness to 1–2/5, hypophonia, and rapidly progressing functional decline. Brain CT showed known mass extending from the brainstem to the thalamus, slightly eccentric on the left. A hemorrhagic component within the mass and slight increase in size of the tumor was noted as well [Figure 4].
Figure 4: Four months after partial resection and shunting; (a) non-contrast brain CT shows hemorrhagic component mostly in the posterior aspect of the tumor; (b) axial, (c) sagittal, and (d) coronal T1 MRI views with contrast, showing significant interval tumor enlargement with solid (solid arrows) and cystic (dotted arrows) components. At this point in the patient’s care, it was felt, he did not have many options left. He appeared to have either poorly tolerated, or did not respond to, chemotherapy. After two prior radiation treatments, he was ineligible for further radiation. Therefore, a multi-disciplinary decision was made to repeat neurosurgical intervention in an attempt to decompress the cyst and remove some more tumor, if feasible.
Second operation
The patient underwent suboccipital-torcular craniotomy with utilization of volumetric image guidance, microscope, and neuromonitoring. The lesion was approached through the supracerebellar infratentorial corridor through the infracollicular safe entry zone. Initially, the cystic portion was decompressed and motor oil-like fluid evacuated. The cyst walls collapsed and the brainstem relaxed. The solid portion of the tumor in the middle of the caudal midbrain and upper pons was considerably softer than in the first surgery, but more vascular with areas of hemorrhagic transformation. This tumor was successfully mobilized and grossly resected with what appeared to be normal appearing brainstem underneath.
Postoperative brain MRI demonstrated GTR and marked decompression of the brainstem. Pathology confirmed mainly necrosis and hemorrhage with focal residual PA with little/no proliferation and Ki-67 in MID1 proliferation indices [Figure 5].
Figure 5: Postoperative brain MRI: (a) Axial T1 image; (b) sagittal T1 image; and (c) coronal T1 image, demonstrating postsurgical changes after GTR of the left dorsal midbrain-pontine tumor.
Postoperative complications
Five days after surgery, the patient experienced increased somnolence. Head CT revealed new bilateral parieto-occipital subdural and epidural retrocerebellar hemorrhage. The parieto-occipital convexity hemorrhages were likely related to intra- and postoperative brain shift and tearing of the bridging veins. Although supratentorial hematomas were not felt to require surgical intervention, the retrocerebellar hemorrhage was more significant. The patient was subsequently taken up for emergent re-opening of the torcular-suboccipital craniotomy for evacuation of the epidural hematoma. Postoperative CT revealed successful evacuation of the extra-axial hematoma.
Three months after surgery, follow-up brain MRI demonstrated evolving postoperative changes with no residual/recurrent tumor.
Twelve months after surgery, the patient’s dysarthria improved. Motor exam revealed persistent right hemiparesis (improved to 3–4/5), arm weaker than the leg. Patient was now able to stand and walk short distances with assistance. Chronic steroids have been successfully weaned off, and the patient had a complete reversal of cushingoid appearance.
DISCUSSION
LGG have been described according to the WHO, as Grades I and II.[7] Among these LGGs, the most common primary brain tumor is PA described as a relatively benign Grade I neoplasm.[7] The most common clinical presentation of brainstem PA is a focal neurological deficit of cranial nerves corresponding to the anatomical location of the tumor with or without involvement of the motor/sensory long tracts.[2] Additional symptoms of hydrocephalus and headache may be present as well.[2] While PAs are considered to have an excellent prognosis with overall 10-year survival reported over 90%, tumors where complete surgical resection is unable to be carried out have less favorable PFS and overall survival.[3,9] In addition, brainstem PA is complicated with significant neurological morbidity secondary to their location.[11]
The gold standard for approaching pediatric brainstem LGGs is initial treatment with safe maximal surgical resection, in conjunction with chemotherapy and radiation therapy as needed.[11] Upadhyaya et al. studied 25 cases of pediatric non tectal brainstem LGGs treated with surgical resection. All three patients with GTR and histological diagnosis of PA demonstrated no evidence of disease at long-term follow-up.[11] Overall, the study found a 10-year PFS rate of 71% in these 25 patients, of which 16 were cases of brainstem PA.[11]
Kestle et al.’s study investigated 28 cases of pediatric brainstem PA, of which 25 were treated with neurosurgical resection.[5] Kestle et al.’s study found a direct correlation between extent of tumor resection and long-term survival.[5] Twelve patients had postoperative imaging demonstrating GTR or linear enhancement, with a 10-year PFS of 62%. In contrast, the 13 patients with postoperative scans revealing residual solid tumor only had a 10-year PFE of 19%.[5] While patients with GTR saw long-term PFS, surgical resection was often associated with neurological dysfunction most apparent in the immediate postoperative phase.[5] This deficit typically resolved but persisted in 6 of the 28 patients.[5]
While GTR of brainstem PA clearly leads to favorable outcomes, adjuvant or neoadjuvant radiation therapy may also be used in the management of these tumors. Gagliardi et al. investigated the use of radiotherapy for LGGs in 39 patients.[4] Of these, four solid lesions converted into mixed tumors following GKRS.[4] In addition, tumor progression was followed in patients and found cyst progression occurred more commonly in patients with a mixed lesion at the time of radiation therapy.[4] Four of five reported incidents of cyst progression were found to be PA.[4] Overall, this study found a 10-year PFS rate of 39.1% in those treated with Gamma Knife Radiosurgery.[4]
The patient in this case report was not offered surgery as the first line of treatment. Instead, he was treated with SRS, another hypofractionated course of radiotherapy, and chemotherapy. It is conceivable that the tumor could acquire its very firm consistency as a consequence of these treatments. Hardening of the tumor due to radiation therapy is a well described phenomenon.[10] Multiple authors reported surgical challenges associated with resection of previously irradiated tumors. Changes in the tumor consistency might be related to the development of fibrosis within the tumor and excessive perifocal scarring. Such changes were noted during the first surgery. However, the second round of chemotherapy with carboplatin and vincristine, along with another cycle with cisplatin and etoposide, was associated with considerable softening of the tumor and its partial hemorrhagic transformation that allowed for gross resection. This phenomenon has not been previously described in the literature. It could be also suggested that between [Figures 3 and 4], a newly enhancing tissue (best seen on the axial and coronal MR images) may represent a growth of radiotherapynaïve tumor, which was softer and thus, more amenable to surgical resection. This hypothesis can be true only if the surgeons did not leave behind a considerable amount of solid tissue after the first surgical procedure, which was not the case. In addition, the tumor, new or old, could outgrow its blood supply and undergo necrotic transformation, which could also be influenced by chemotherapy. It is important to note, though, that the new radiotherapy-naïve component was exposed to chemotherapy, much like the hard tumor treated with XRT. At this point, we can only speculate about the extent to which the above potential mechanisms could have contributed to the apparent softening of the mass. Although not unequivocally proving the exact causality, the facts that remain clear are (1) very hard consistency of the tumor discovered on the first surgery and (2) a quite significant softening of previously unresectable lesion after chemotherapy. Therefore, the authors believe that chemotherapy could have contributed to the success of the second surgery.
We did not find any reports of repeat surgical interventions on adult patients with brainstem LGG. One study reported 12 pediatric patients with midbrain LGG who were treated with multiple repeat tumor resections due to disease progression with seemingly stable long-term results obtained in at least nine patients.[8] The surgeons treating brainstem LGG should be aware of possibility of chemotherapy induced tumor softening phenomenon and be prepared to offer repeat surgery, even if the lesion was initially deemed unresectable.
CONCLUSION
Aggressive treatment of brainstem LGG should be pursued whenever possible, given its generally favorable prognosis. Repeat microsurgical resection, even with a different approach, might be reasonable and safe. Finally, chemotherapy may be associated with changes in the tumor consistency that can render previously unresectable lesions amenable to successful aggressive resection.
How to cite this article: Chung DJ, Arif B, Odia Y, Siomin V. Chemotherapy-induced changes in tumor consistency can allow gross total resection of previously unresectable brainstem pilocytic astrocytoma. Surg Neurol Int 2021;12:12.
Declaration of patient consent
The authors certify that they have obtained all appropriate patient consent.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest. | Oral | DrugAdministrationRoute | CC BY-NC-SA | 33500827 | 19,408,198 | 2021 |
What was the outcome of reaction 'Seizure'? | Chemotherapy-induced changes in tumor consistency can allow gross total resection of previously unresectable brainstem pilocytic astrocytoma.
Low-grade gliomas (LGG) are described by the World Health Organization as Grades I and II. Among LGGs, the most common primary brain tumor is pilocytic astrocytoma (PA) and carries an excellent prognosis when treated with complete surgical resection. Cases, in which this is not possible, are associated with less favorable outcomes and worse progression-free survival.
This report describes a case of a 22-year-old male, who presented with progression of a primary brainstem tumor previously treated with stereotactic radiosurgery and chemotherapy. Patient underwent surgical exploration and was diagnosed with juvenile PA, but debulking was limited by the very dense and fibrous tumor. Complete surgical resection was not possible at this time. Despite efforts to treat with chemotherapy, the patient presented a year later with clinical deterioration and severe neurologic deficits, prompting surgical re-exploration. During the second operation, the tumor was found to have undergone very significant softening in consistency, allowing for gross total resection (GTR).
Aggressive treatment of brainstem LGG should be pursued whenever possible, given its generally favorable prognosis. Repeat microsurgical resection, even with a different approach, might be reasonable and safe. Finally, chemotherapy may be associated with changes in the tumor consistency that can render previously unresectable lesions amenable to successful aggressive resection.
INTRODUCTION
Low-grade gliomas (LGG) have been described according to the World Health Organization (WHO), as Grades I and II.[7] Among these LGGs, the most common primary brain tumor is PA, accounting for 5.1% of all primary brain and central nervous system gliomas, and are described as relatively benign Grade I neoplasms.[7,9] It is the most prevalent primary brain tumor among children age 0–19 years, representing 15.3% of all cases.[9] Among children age 0–14 years, PAs have an annual age-adjusted incidence rate of 1.01 (per 100,000) that significantly decreases (0.28/100,000) in adolescents and young adults (age 15–30 years).[9] PAs in children most commonly occur in the cerebellum, optic pathway/ hypothalamic region, and supratentorial hemispheres but less frequently may occur in the brainstem and spinal cord.[2] PAs commonly arise sporadically but there is an increased risk occurrence in the optic pathway in patients with neurofibromatosis Type 1.[3]
PAs are typically well-defined tumors with cystic formation occurring within the tumor or around the tumor with associated solid nodule.[3] Histopathologically, PAs demonstrate a biphasic growth pattern with areas of highly fibrillated cells mixed with loose microcystic component accompanied by Rosenthal fibers and eosinophilic granular bodies.[3] Microvascular proliferation is common but does not indicate malignancy like in diffuse high-grade gliomas.[3] Recently, genetic alterations in the mitogen-activated protein kinase pathway of PAs have been described in literature with the most common alteration being a fusion between KIAA1549 and BRAF.[3]
While PAs are considered to have an excellent prognosis with overall 10-year survival reported over 90%, tumors where complete surgical resection is unable to be carried out have less favorable progression-free survival (PFS) and overall survival.[3,9] Primary treatment of these tumors is surgery, but radiotherapy and chemotherapy may be indicated in cases with incomplete resection or contraindication to surgery.[3] Among reported brainstem LGGs treated with surgical resection or chemotherapy, there is a clear positive correlation with extent of surgical resection and PFS [Table 1]. For example, Ahmed et al. describes 21 patients who received gross total resection (GTR) or subtotal resection (STR) with 5-year overall survival of 85%, while Klimo et al. reported 9 patients who received GTR with 5-year PFS at 75%.[1, 6]
Table 1: Summary of previously reported brainstem low-grade glioma managed with surgical resection or chemotherapy.
However, the repeat microsurgical resection of partially resected tumors that have been treated with chemotherapy resulting in gross total resection (GTR) has not been previously reported. Here, we present a case of a 22-year-old male with an initially unresectable brainstem LGG treated with vincristine and carboplatin, subsequently undergoing changes in consistency allowing for GTR on repeat surgery.
CASE PRESENTATION
History
On March 2015, a 22-year-old right-handed male with a 2 month history of right sided weakness and numbness was diagnosed with a primary brainstem tumor on MRI. He was empirically treated with temozolomide, bevacizumab, and stereotactic radiosurgery (SRS). Six months later, while stable disease was noted on brain MRI, the patient was also treated with an additional hypo-fractionated course (~50Gy in 25 fractions over 2 months) of involved field radiation.
Patient received empiric chemotherapy and radiation therapy for 21 months and follow-up brain MRI on December 2016 revealed a stable, necrotic left midbrain mass lesion measuring 2.5 × 2.1 × 1.8 cm with subacute hemorrhage in the inferior posterior margin [Figure 1].
Figure 1: Initial brain MRI: (a) Axial T1 image with contrast; (b) coronal FLAIR and (c) coronal T1 with contrast images, showing large left midbrain-pontine lesion at the time of stereotactic radiosurgery.
Nearly 2 years after the initial diagnosis, the patient presented to the clinic with progressively worsening gait instability and left greater than right weakness. Additional symptoms included transient confusion with malaise and dizziness. Brain MRI at this time revealed progression at medial and inferior margins along with interval mild ventriculomegaly consistent with obstructive hydrocephalus [Figure 2]. Decision was made to place the shunt and resect the tumor in a two-stage procedure during the same admission.
Figure 2: Two years after stereotactic radiosurgery and first round of chemotherapy: (a) Brain CT showing hemorrhage in the dorsal aspect of the tumor. Brain MRI with contrast: (b) Axial T1 image demonstrating the lesion in the left brainstem and prominent temporal horns (arrows), suggestive of an obstructive hydrocephalus; (c) sagittal T1 image showing ring-enhancing mass.
First operation – VP shunt placement and partial tumor resection/biopsy
The patient underwent an uneventful right occipital ventriculoperitoneal shunt placement with utilization of volumetric image guidance and laparoscopic assistance.
The following day, left posterior temporal-occipital craniotomy was performed with an intention to obtain tissue for diagnosis and remove as much of the brainstem tumor as safely as possible. Volumetric image guidance, microscope, neuro-monitoring, and intraoperative MRI were utilized. The brainstem lesion was approached through the posterior temporal-occipital supratentorial plane with sectioning of the tentorium to widen the exposure. The inferior posterior temporal gyrus was partially resected, allowing mobilization and preservation of the vein of Labbe. The midbrain was entered through the lateral mesencephalic safe entry zone. The tumor itself was very firm and moderately vascular. Usual microsurgical tools, including ultrasonic aspirator, were inefficient.
The tumor’s firm consistency did not allow us to remove much of the lesion. The outer portions of the mass were very fibrous and could not be mobilized. Progress was very slow. The remaining tumor was “hard as a rock” in consistency, and surgical manipulation would move it as a single block, distorting the entire brainstem. Considering the circumstances, it was felt that the safest choice would be to stop further dissection/tumor removal. Intraoperative MRI showed an approximately 40% decrease in tumor volume, but the peripherally enhancing capsule of the tumor remained unchanged, as expected given the consistency of the tumor encountered intraoperatively [Figure 3].
Figure 3: MRI images with contrast: (a) Immediate preoperative axial T1 views; (b) intraoperative Axial T1 views demonstrate an approximate 40% resection of the lesion (dotted arrow) and expected intraoperative pneumocephalus (solid arrows); (c) immediate preoperative coronal T1 views; (d) intraoperative coronal T1 views.
Pathology and postoperative course
Final pathology confirmed PA, negative for BRAF mutation or rearrangement, and the patient was started on combination therapy with carboplatin and vincristine. Unfortunately, chemotherapy was complicated with breakthrough seizures and brain MRI revealed symptomatic interval progression of the left midbrain tumor. Carboplatin and vincristine were discontinued, and seizures controlled with levetiracetam, gabapentin, and valproic acid. The patient was subsequently started on a 3 day monthly cycle of cisplatin and etoposide.
On May 2018, 14 months after the first operation, the patient presented to the emergency department for worsening headache, double vision, right-sided weakness, and gait ataxia. Clinically, he was noted to have significant right upper extremity weakness to 1–2/5, hypophonia, and rapidly progressing functional decline. Brain CT showed known mass extending from the brainstem to the thalamus, slightly eccentric on the left. A hemorrhagic component within the mass and slight increase in size of the tumor was noted as well [Figure 4].
Figure 4: Four months after partial resection and shunting; (a) non-contrast brain CT shows hemorrhagic component mostly in the posterior aspect of the tumor; (b) axial, (c) sagittal, and (d) coronal T1 MRI views with contrast, showing significant interval tumor enlargement with solid (solid arrows) and cystic (dotted arrows) components. At this point in the patient’s care, it was felt, he did not have many options left. He appeared to have either poorly tolerated, or did not respond to, chemotherapy. After two prior radiation treatments, he was ineligible for further radiation. Therefore, a multi-disciplinary decision was made to repeat neurosurgical intervention in an attempt to decompress the cyst and remove some more tumor, if feasible.
Second operation
The patient underwent suboccipital-torcular craniotomy with utilization of volumetric image guidance, microscope, and neuromonitoring. The lesion was approached through the supracerebellar infratentorial corridor through the infracollicular safe entry zone. Initially, the cystic portion was decompressed and motor oil-like fluid evacuated. The cyst walls collapsed and the brainstem relaxed. The solid portion of the tumor in the middle of the caudal midbrain and upper pons was considerably softer than in the first surgery, but more vascular with areas of hemorrhagic transformation. This tumor was successfully mobilized and grossly resected with what appeared to be normal appearing brainstem underneath.
Postoperative brain MRI demonstrated GTR and marked decompression of the brainstem. Pathology confirmed mainly necrosis and hemorrhage with focal residual PA with little/no proliferation and Ki-67 in MID1 proliferation indices [Figure 5].
Figure 5: Postoperative brain MRI: (a) Axial T1 image; (b) sagittal T1 image; and (c) coronal T1 image, demonstrating postsurgical changes after GTR of the left dorsal midbrain-pontine tumor.
Postoperative complications
Five days after surgery, the patient experienced increased somnolence. Head CT revealed new bilateral parieto-occipital subdural and epidural retrocerebellar hemorrhage. The parieto-occipital convexity hemorrhages were likely related to intra- and postoperative brain shift and tearing of the bridging veins. Although supratentorial hematomas were not felt to require surgical intervention, the retrocerebellar hemorrhage was more significant. The patient was subsequently taken up for emergent re-opening of the torcular-suboccipital craniotomy for evacuation of the epidural hematoma. Postoperative CT revealed successful evacuation of the extra-axial hematoma.
Three months after surgery, follow-up brain MRI demonstrated evolving postoperative changes with no residual/recurrent tumor.
Twelve months after surgery, the patient’s dysarthria improved. Motor exam revealed persistent right hemiparesis (improved to 3–4/5), arm weaker than the leg. Patient was now able to stand and walk short distances with assistance. Chronic steroids have been successfully weaned off, and the patient had a complete reversal of cushingoid appearance.
DISCUSSION
LGG have been described according to the WHO, as Grades I and II.[7] Among these LGGs, the most common primary brain tumor is PA described as a relatively benign Grade I neoplasm.[7] The most common clinical presentation of brainstem PA is a focal neurological deficit of cranial nerves corresponding to the anatomical location of the tumor with or without involvement of the motor/sensory long tracts.[2] Additional symptoms of hydrocephalus and headache may be present as well.[2] While PAs are considered to have an excellent prognosis with overall 10-year survival reported over 90%, tumors where complete surgical resection is unable to be carried out have less favorable PFS and overall survival.[3,9] In addition, brainstem PA is complicated with significant neurological morbidity secondary to their location.[11]
The gold standard for approaching pediatric brainstem LGGs is initial treatment with safe maximal surgical resection, in conjunction with chemotherapy and radiation therapy as needed.[11] Upadhyaya et al. studied 25 cases of pediatric non tectal brainstem LGGs treated with surgical resection. All three patients with GTR and histological diagnosis of PA demonstrated no evidence of disease at long-term follow-up.[11] Overall, the study found a 10-year PFS rate of 71% in these 25 patients, of which 16 were cases of brainstem PA.[11]
Kestle et al.’s study investigated 28 cases of pediatric brainstem PA, of which 25 were treated with neurosurgical resection.[5] Kestle et al.’s study found a direct correlation between extent of tumor resection and long-term survival.[5] Twelve patients had postoperative imaging demonstrating GTR or linear enhancement, with a 10-year PFS of 62%. In contrast, the 13 patients with postoperative scans revealing residual solid tumor only had a 10-year PFE of 19%.[5] While patients with GTR saw long-term PFS, surgical resection was often associated with neurological dysfunction most apparent in the immediate postoperative phase.[5] This deficit typically resolved but persisted in 6 of the 28 patients.[5]
While GTR of brainstem PA clearly leads to favorable outcomes, adjuvant or neoadjuvant radiation therapy may also be used in the management of these tumors. Gagliardi et al. investigated the use of radiotherapy for LGGs in 39 patients.[4] Of these, four solid lesions converted into mixed tumors following GKRS.[4] In addition, tumor progression was followed in patients and found cyst progression occurred more commonly in patients with a mixed lesion at the time of radiation therapy.[4] Four of five reported incidents of cyst progression were found to be PA.[4] Overall, this study found a 10-year PFS rate of 39.1% in those treated with Gamma Knife Radiosurgery.[4]
The patient in this case report was not offered surgery as the first line of treatment. Instead, he was treated with SRS, another hypofractionated course of radiotherapy, and chemotherapy. It is conceivable that the tumor could acquire its very firm consistency as a consequence of these treatments. Hardening of the tumor due to radiation therapy is a well described phenomenon.[10] Multiple authors reported surgical challenges associated with resection of previously irradiated tumors. Changes in the tumor consistency might be related to the development of fibrosis within the tumor and excessive perifocal scarring. Such changes were noted during the first surgery. However, the second round of chemotherapy with carboplatin and vincristine, along with another cycle with cisplatin and etoposide, was associated with considerable softening of the tumor and its partial hemorrhagic transformation that allowed for gross resection. This phenomenon has not been previously described in the literature. It could be also suggested that between [Figures 3 and 4], a newly enhancing tissue (best seen on the axial and coronal MR images) may represent a growth of radiotherapynaïve tumor, which was softer and thus, more amenable to surgical resection. This hypothesis can be true only if the surgeons did not leave behind a considerable amount of solid tissue after the first surgical procedure, which was not the case. In addition, the tumor, new or old, could outgrow its blood supply and undergo necrotic transformation, which could also be influenced by chemotherapy. It is important to note, though, that the new radiotherapy-naïve component was exposed to chemotherapy, much like the hard tumor treated with XRT. At this point, we can only speculate about the extent to which the above potential mechanisms could have contributed to the apparent softening of the mass. Although not unequivocally proving the exact causality, the facts that remain clear are (1) very hard consistency of the tumor discovered on the first surgery and (2) a quite significant softening of previously unresectable lesion after chemotherapy. Therefore, the authors believe that chemotherapy could have contributed to the success of the second surgery.
We did not find any reports of repeat surgical interventions on adult patients with brainstem LGG. One study reported 12 pediatric patients with midbrain LGG who were treated with multiple repeat tumor resections due to disease progression with seemingly stable long-term results obtained in at least nine patients.[8] The surgeons treating brainstem LGG should be aware of possibility of chemotherapy induced tumor softening phenomenon and be prepared to offer repeat surgery, even if the lesion was initially deemed unresectable.
CONCLUSION
Aggressive treatment of brainstem LGG should be pursued whenever possible, given its generally favorable prognosis. Repeat microsurgical resection, even with a different approach, might be reasonable and safe. Finally, chemotherapy may be associated with changes in the tumor consistency that can render previously unresectable lesions amenable to successful aggressive resection.
How to cite this article: Chung DJ, Arif B, Odia Y, Siomin V. Chemotherapy-induced changes in tumor consistency can allow gross total resection of previously unresectable brainstem pilocytic astrocytoma. Surg Neurol Int 2021;12:12.
Declaration of patient consent
The authors certify that they have obtained all appropriate patient consent.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest. | Recovering | ReactionOutcome | CC BY-NC-SA | 33500827 | 19,382,011 | 2021 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Atrial septal defect'. | Ivabradine use in pregnant women-treatment indications and pregnancy outcome: an evaluation of the German Embryotox database.
OBJECTIVE
Ivabradine has been approved for the treatment of chronic heart failure and chronic stable angina pectoris in Europe. Based on adverse outcomes of reproductive animal studies and the lack of human data, ivabradine is considered contraindicated during pregnancy. The aim of this observational study is to analyse ivabradine use before and during pregnancy.
METHODS
We evaluated all ivabradine-related requests to the German Embryotox Institute from 2007 to 2019. Exposed pregnancies were analysed as to their outcome.
RESULTS
Off-label use for supraventricular tachycardia was frequent in women of childbearing age. Of 38 prospectively ascertained pregnancies with ivabradine exposure and completed follow-up, 32 resulted in live births, 3 in spontaneous abortions, and 3 were electively terminated. One neonate presented with major birth defects (atrial septal defect and cleft palate). In 33/38 patients, ivabradine was discontinued after confirmation of pregnancy without cardiac deterioration and 5/38 women continued ivabradine throughout pregnancy. In addition, there were 3 retrospectively reported pregnancies including one major birth defect (tracheal atresia).
CONCLUSIONS
This case series represents the largest cohort of ivabradine-exposed pregnancies, published so far. According to our findings, ivabradine appears not to be a major teratogen. However, established drugs of choice with strong evidence of low risk for the unborn should be preferred in women planning pregnancy. After inadvertent exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, close monitoring is necessary in pregnant women with supraventricular arrhythmias or cardiac disease.
Introduction
Ivabradine is a heart rate-lowering agent, acting through selective inhibition of the If (I-funny) channel without additional hemodynamic effects on cardiac function. The approved treatment indications in Europe are chronic heart failure and symptomatic treatment of chronic stable angina pectoris after failing or intolerance of beta-blockers [1].
Preclinical studies have suggested fetotoxicity and teratogenic effects. In pregnant rats, ivabradine exposure equivalent to therapeutic levels in humans was associated with increased intrauterine and postnatal mortality. Ventricular septal defects and complex anomalies of the great arteries were noted at doses 3 times the therapeutic human exposure. In pregnant rabbits, reduced fetal and placental weight as well as ectrodactyly was observed at doses 15 times the therapeutic human exposure [2]. In addition, dose-dependent mortality was seen in chicken and mice embryos [3, 4].
ESC guidelines consider ivabradine contraindicated during pregnancy and recommend discontinuation independent of treatment indication before conception with close clinical and echocardiographic monitoring [5].
Published data on ivabradine-exposed pregnancies are limited to three case reports, one with ivabradine treatment in the late first and early second trimester for sinus tachycardia after myocardial infarction [6], and the second starting ivabradine in the second trimester for deteriorating tachycardia-induced cardiomyopathy [7]. In the third case with first trimester exposure, pregnancy was electively terminated [8].
The limited human experience along with the teratogenic effects in animal studies makes it difficult to counsel inadvertently exposed pregnant women and their health care providers (HCP). Therefore, we decided to evaluate ivabradine-exposed pregnancies recorded by the German Embryotox Institute.
Methods
Study design
The German Embryotox Institute offers risk assessment on drug exposure during pregnancy to HCPs and patients. Up to 15,000 annual requests are answered, and in approximately 3500 critically exposed cases per year, the course and outcome of pregnancy are documented. At the initial contact, maternal characteristics and detailed exposure assessment are asked for after informed consent. About 8 weeks after the estimated date of birth, information on course and outcome of pregnancy is collected via structured telephone interview and mailed questionnaires. After a case by case plausibility check, additional information and health care records are requested in cases of adverse outcomes or inconsistent information.
Weeks of pregnancy were based on ultrasound and if not available on the last menstrual period (LMP). First trimester was defined as gestational week 2 + 0 (conception) until 12 + 6. Major birth defects were classified according to EUROCAT [9]. In prospectively ascertained pregnancies, neither the outcome of pregnancy nor prenatal pathology was known at the time of first contact. Pregnancies that were reported after birth or because of prenatal pathology were considered retrospective and evaluated separately. A detailed description of methodology adapted to the recommendations of the Strengthening of the Reporting of Observational studies in Epidemiology (STROBE) statement is given in Schaefer et al. [10] and Dathe and Schaefer [11].
We evaluated all requests on ivabradine to the German Embryotox Institute from 2007 to 2019.
Statistical analyses
Descriptive statistics were applied. For age, BMI, gestational weeks, neonatal weight, length, and head circumference, median and interquartile ranges are presented. Birth weights were adjusted to sex and gestational age at birth and percentile categories and standard deviation scores (SDS) were calculated according to the German perinatal survey [12].
Results
During the study period 2007–2019, our institute received 97 requests for ivabradine, 56 (58%) were from HCP, and 41 from patients. For further details, see Fig. 1. Treatment indications for ivabradine over time in relation to formal approval status and pregnancy labelling are summarized in Fig. 2. Supraventricular tachycardia was the most common reason for ivabradine treatment in our cohort. The first ivabradine-exposed pregnancy was recorded in March 2007, 1 year after approval and marketing authorization in Germany in January 2006.Fig. 1 Overview on information requests to the German Embryotox institute on ivabradine
Fig. 2 Treatment indication for ivabradine in 97 Embryotox requests over time in relation to pregnancy labelling and approval status by EMA/FDA. Recommendations for use of ivabradine in supraventricular tachycardias, i.e. inappropriate sinus tachycardia and postural orthostatic tachycardia syndrome: HRS/ESC 2015 [13], ACC/AHA/HFSA 2016 [14], ESC 2019 [15]. ACC, American College of Cardiology; EMA, European Medicines Agency; ESC, European Society of Cardiology; FDA, US Food and Drug Administration; HRS, Heart Rhythm Society
Fifty-six of the 97 requests were related to maternal exposure during pregnancy (Fig. 1). In 41 of these cases, follow-up on pregnancy outcome could be completed: 38 pregnancies were prospectively ascertained and three were retrospective reports.
Evaluation of prospectively ascertained pregnancies
Maternal characteristics
The median age of pregnant women was 29.5 years and the median BMI 22.5. Maternal characteristics are presented in detail in Table 1. The mean time at first contact was gestational week (GW) 7. Only 16 of the 38 prospectively ascertained pregnancies were actually planned (42%) and only 7 women (18%) started folic acid before conception. Contraceptive methods, mostly oral contraceptives (n = 4), were still used by 6 women at conception.Table 1 Maternal characteristics and pregnancy outcomes of ivabradine exposed pregnancies. Data are n (%) except for age, BMI, gestational week at birth, neonatal weight, length and head circumference, which are median and interquartile range
Maternal characteristics N (%) or median (IQR)
Maternal characteristics Age (n=38) 29.5 (26.5-34)
BMI (n=34) 22.5 (19.4-25.6)
Smoking (n=37)
>5 cigarettes/day 7 (19%)
<=5 cigarettes/day 4 (11%)
Alcohol (n=36)
<=1 drink/day 2 (6%)
>1 drink/day 1 (3%)
GW at first contact (n=38) 7 (5.6-11.1)
Previous pregnancies (n=38)
0 17(45%)
1 8 (21%)
2 13 (34%)
Pregnancy outcome and neonatal characteristics N (%) or median (IQR)
Pregnancy outcome (n=38) Spontaneous abortion 3
ETOP 3
Live birth 32
Pregnancy complications (n=32, only live births) Gestational diabetes 5 (16%)
Pre-eclampsia 2 (6%)
Cesarean section 15 (46%)
Preterm birth (<GW 37) 3 (9%)
Neonatal characteristics (n=33, including one pair of twins) GW at birth (n=32) 39 (37.9-39.9)
Weight, g (n=32) 3300 (2935-3610)
Length, cm (n=31) 50 (49-52.5)
HC, cm (n=30) 35 (34-36)
SGA (n=32) 2 (6%)
BMI, Body Mass Index; ETOP, elective termination of pregnancy; GW, gestational week; HC, head circumference; IQR, interquartile range; SGA, small for gestational age (<10th percentile)
Ivabradine exposure and co-medication
The most prevalent indication for ivabradine treatment was supraventricular tachycardia (n = 29) followed by heart failure (n = 3) and coronary artery disease (n = 2). Other reported indications (n = 4) included QT prolongation and sick-sinus-syndrome, although both are regarded as contraindications for ivabradine. The median daily ivabradine dose was 7.5 mg (IQR 5–10, min–max 2.5–15, n = 33).
Nearly all women were treated with ivabradine during the first trimester (n = 37) and 32 stopped treatment after recognition of pregnancy (Fig. 3), at median gestational week 6.1 (interquartile range 5.4–7.8). Cardiac decompensation or complications were not described after cessation of treatment. Only 5 women continued ivabradine throughout pregnancy or until shortly before delivery. One woman started treatment after first trimester (GW 18) of her twin pregnancy due to symptomatic tachycardia under beta-blocker therapy (Iv34 in Fig. 3).Fig. 3 Ivabradine exposure and pregnancy outcome of prospectively (n = 38) and retrospectively (n = 3) ascertained pregnancies
Cardiovascular co-medication consisted of beta-blockers (mainly metoprolol) in 15 women, mostly concomitantly (n = 7) or in replacement of ivabradine (n = 6). Two women initiated ivabradine after discontinuation of beta-blockers (Iv30 and Iv34 in Fig. 3). Six women received other cardiovascular medication, three methyldopa or amlodipine for hypertension, two digitoxin or verapamil for tachyarrhythmia, and one sildenafil and eplerenone for cardiomyopathy with cardiac failure (Iv17).
Psychiatric co-medication was reported in 7/38 women (5 of these with an anxiety disorder) including the teratogenic antiepileptic topiramate (n = 1) and pregabalin (n = 3).
Pregnancy outcome
Of the 38 prospectively ascertained ivabradine-exposed pregnancies, 32 were live births (33 infants including one pair of twins) and three resulted in early spontaneous abortions (Iv10, Iv17, and Iv19 in Fig. 3). One of these occurred in a woman with cardiomyopathy (Iv17) and various co-medications including hydroxycarbamide for thrombocythemia until GW 7. Three pregnancies were electively terminated for social/psychological reasons. Further information on pregnancy complications and outcomes is given in Table 1 and Fig. 3.
Major birth defects were reported in one pregnancy (Iv24) with 15 mg ivabradine plus bisoprolol until GW 7 + 5, both replaced by metoprolol. Polyhydramnios and preeclampsia led to Caesarean section in GW 35 after premature rupture of membranes. The premature child was large for gestational age (> 97 percentile) and had a cleft palate, an atrial septal defect II (spontaneously closed at the age of 10 months), and a hypertrophic cardiac septum. The mother was treated for sinus tachycardia, in addition she had diabetes type I and a BMI of 30.
Neonatal characteristics
Characteristics of the 33 neonates are summarized in Table 1. Compared to the German perinatal survey, standardized birth weights were lower with a median SDS of − 0.36, corresponding to a reduction of birth weight of 149 g for singletons born in gestational week 40.
Two neonates were small for gestational age, one of the twins (exposure only during 2nd and 3rd trimester, Iv34) and one after maternal ivabradine exposure until GW 12 (Iv29). Two neonates were large for gestational age, both mothers were diabetic (Iv24, type I diabetes and Iv27, gestational diabetes).
Retrospective cases
During the study period three pregnancies were retrospectively reported (further details are given in Table 2 and Fig. 3). One premature neonate (IvR1) with tracheal atresia died perinatally. The mother had been treated with ivabradine, naproxen, and metamizole during the first trimester until recognition of pregnancy (no further details available).Table 2 Retrospectively reported pregnancies
Case Maternal age/BMI Treatment indication Ivabradine exposure, other co-medication (GW or trimester) Pregnancy outcome, GW, birth weight percentile, sex Pregnancy complications, birth defects, and additional findings
IvR1 31 years/31.2 Tachycardia Ivabradine (1st trim)
Naproxen (1st trim)
Metamizole (1st trim)
Live birth, 36 + 2, > 97th, f Polyhydramnios, perinatal death, tracheal atresia, normal karyotype (46, XX)
IvR2 29 years/18.4 Tachycardia, syncopes, pacemaker Ivabradine (1st trim)
Bisoprolol (0–35 + 5)
Levetiracetam (0–35 + 5, grand mal epilepsy)
Live birth, 35 + 5, < 10th, f IUGR (since GW 30), oligohydramnios (since GW 34), Caesarean section (vaginal bleeding)
IvR3 29 years/not available Tachycardia Ivabradine (0–41 + 3)
Bisoprolol (0–41 + 3)
Live birth, 41 + 3
> 10th, m
Uncomplicated pregnancy
BMI, body mass index; f, female; GW, gestational week; IUGR, intrauterine growth restriction; m, male; trim, trimester
The other two retrospective cases include a woman with comorbidities, intrauterine growth restriction, and premature birth (IvR2). The second was a healthy newborn (IvR3) after treatment with ivabradine throughout pregnancy. This mother reported a previous also uneventful pregnancy course under ivabradine.
Discussion
Pregnancy outcome
In our prospective cohort, there was only one neonate with major birth defects (cleft palate and atrial septal defect) from a diabetic mother with supraventricular tachycardia. No further major or cardiac malformations were reported among the 32 live births and 6 pregnancy losses. Cardiac defects have been observed in preclinical experimental studies. As a possible teratogenic mechanism, ivabradine-induced embryonic heart rate reduction leading to reduced cardiac output and hypoxia was discussed [16]. With a prevalence of almost 1 of 100 cardiac malformations belong to the most frequent birth defects [17, 18]. In our case, the neonate was large for gestational age and had a septum hypertrophy, supporting the assumption that maternal pre-gestational diabetes may also be causative for the observed cardiac defect and cleft palate. A causal relationship between ivabradine and cleft palate is unlikely, because ivabradine was discontinued at GW 7 + 5 before the beginning of the critical time period for the development of cleft palate at gestational week 9. The rate of major birth defects in our prospective cohort is comparable to the expected background rate of 2–3% [17, 18]. However, the limited number of pregnancies exposed during the entire period of embryogenesis does not allow excluding an increased risk, so far.
There was only one retrospectively reported case with a birth defect (malformation of the trachea) suspected as potentially ivabradine related. The concomitant exposures with naproxen and metamizole have not been evidenced as teratogens in human, whereas the high maternal BMI in the present case poses a risk for birth defects.
Taken together, we did not observe a specific pattern of congenital anomalies in our study. In addition, the presence of maternal comorbidities like diabetes and high BMI may have contributed to the observed birth defects.
The number of spontaneous abortions (3/38) was not higher than expected. The rate of preeclampsia and preterm birth in our cohort was within the expected range [19].
Treatment indications
Although the approved treatment indications for ivabradine, chronic heart failure and chronic stable angina pectoris (see Fig. 2 for changes of licenced use during the study period), are rare in women of reproductive age, we have observed an increasing need for information for off-label use in (younger) women considering pregnancy.
In our case series, most women were treated (off-label) for supraventricular tachycardia using lower doses than recommended (5–7.5 mg twice daily). Supraventricular tachycardia is not uncommon, especially in younger women, and is expected in 0.02–0.5% of pregnancies [20]. Aggravation of symptoms during pregnancy is described in about 20–50% of women related to various mechanisms, including pregnancy-related increase in cardiac heart rate and output and hormonal changes [20].
In 2015, the HRS (Heart Rhythm Society) considered ivabradine as promising for treating patients with inappropriate sinus tachycardia (IST) [13], followed by a recommendation for ivabradine by the ACC (American College of Cardiology) in 2016 [14]—though only one small randomized controlled study has yet supported beneficial effects [21]. Although still not approved for this indication, recent ESC guidelines [15] also recommend ivabradine alone or in combination with a beta-blocker in symptomatic patients with IST, for the treatment of focal atrial tachycardia or for postural orthostatic tachycardia syndrome (POTS), a condition considered as common cause for orthostatic intolerance in women between 15 and 25 years [20, 22] (Fig. 2).
Beta-blockers were required in 18% of women in our prospective cohort (Fig. 3). Discontinuation of ivabradine or switching to beta-blockers after recognition of pregnancy was well tolerated in most of our patients. Thus, necessity of ivabradine treatment should be critically considered in women with supraventricular tachycardia planning pregnancy. Data from previous studies suggest that discontinuation of ivabradine is possible in many (non-pregnant) patients with IST without recurrence. Even beneficial long-term effects after cessation of therapy are discussed [23, 24].
A high number of women in our cohort received therapy for anxiety and/or other psychiatric conditions. This is in line with findings from other studies reporting high rates of anxiety (25%) and depression (26%) in IST patients [25]. Anxiety is considered as non-cardiac symptom of inappropriate sinus tachycardia, and vice versa, anxiety may trigger tachycardia. Especially during pregnancy, empathic care and reassurance are necessary, to prevent increased concerns about possible harm to the fetus [26].
Preconception counselling is of utmost importance in all women considering pregnancy under medication [27]. Sinus tachycardia is a condition diagnosed mainly in young females and the first diagnosis is frequently made during or after pregnancy [25]. Considering the increasing off-label use of ivabradine as well as the high rate of > 40% unplanned pregnancies [28], more exposed pregnancies can be expected in the future. Well-established treatment options such as selected beta-blockers with low reproductive risk [29] should be considered in women of childbearing age before prescription of ivabradine.
Strength and limitations
A major strength of our study is the detailed exposure protocol and the multi-source ascertainment of pregnancy outcome via patients and their HCP followed by a case by case plausibility test.
Our data may not be representative of the German pregnant population; insofar, women with higher education are over-represented [30]. However, the regional distribution of enrolled patients is representative of the female population of childbearing age in Germany [11]. In addition, women on long-term ivabradine with severe cardiovascular conditions such as the approved treatment indications coronary artery disease and cardiac failure are under-represented in our cohort possibly favouring uneventful pregnancy outcomes. However, given the rarity of severe cardiac conditions in pregnant women, large study cohorts are not to be expected in the foreseeable future.
Conclusions
Our findings do not indicate that ivabradine is a major teratogen when used in early pregnancy. Although being the largest study published so far on the outcome of ivabradine-exposed pregnancies, the sample is too small to rule out embryotoxic effects. Further studies are needed to confirm or refute our findings.
In women planning pregnancy, established drugs of choice with low risk for the unborn should be preferred to ivabradine. For women with significant arrhythmias and severe cardiac disease, the impact of the underlying medical condition including comorbidities such as diabetes and high BMI on pregnancy outcome has to be considered when planning pregnancy.
After inadvertent ivabradine exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, all pregnant women with complicated arrhythmias or major cardiac disease should be carefully monitored.
Acknowledgements
We would like to thank all patients and physicians reporting pregnancies to the institute and acknowledge our colleagues from the German Embryotox Institute for counselling and thoroughly documenting cases, especially Sandra Zinke for continuous support with case documentation.
Contributions of authors
MH, ML, and CS designed the study. ML, EB, MH, and KD validated and analysed the data. ML, MH, and CS wrote the first draft of the manuscript, and all the authors critically revised subsequent manuscript drafts and contributed essential discussion points. All the authors approved the final version of this manuscript and they are responsible for the accuracy of this work.
Funding
Open Access funding enabled and organized by Projekt DEAL. This work was performed with financial support from the German Ministry of Health (BMG) and the German Federal Institute for Drugs and Medical Devices (BfArM). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
The study was registered with the German Clinical Trial register (DRKS00010502) and the study protocol was approved by the ethics committee of the Charité-Universitätsmedizin Berlin (EA1/107/16).
The study was registered with the German Clinical Trial register (DRKS00010502, 20.5.2016)
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | BISOPROLOL, IVABRADINE, METOPROLOL | DrugsGivenReaction | CC BY | 33501507 | 19,697,614 | 2021-07 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Cardiac hypertrophy'. | Ivabradine use in pregnant women-treatment indications and pregnancy outcome: an evaluation of the German Embryotox database.
OBJECTIVE
Ivabradine has been approved for the treatment of chronic heart failure and chronic stable angina pectoris in Europe. Based on adverse outcomes of reproductive animal studies and the lack of human data, ivabradine is considered contraindicated during pregnancy. The aim of this observational study is to analyse ivabradine use before and during pregnancy.
METHODS
We evaluated all ivabradine-related requests to the German Embryotox Institute from 2007 to 2019. Exposed pregnancies were analysed as to their outcome.
RESULTS
Off-label use for supraventricular tachycardia was frequent in women of childbearing age. Of 38 prospectively ascertained pregnancies with ivabradine exposure and completed follow-up, 32 resulted in live births, 3 in spontaneous abortions, and 3 were electively terminated. One neonate presented with major birth defects (atrial septal defect and cleft palate). In 33/38 patients, ivabradine was discontinued after confirmation of pregnancy without cardiac deterioration and 5/38 women continued ivabradine throughout pregnancy. In addition, there were 3 retrospectively reported pregnancies including one major birth defect (tracheal atresia).
CONCLUSIONS
This case series represents the largest cohort of ivabradine-exposed pregnancies, published so far. According to our findings, ivabradine appears not to be a major teratogen. However, established drugs of choice with strong evidence of low risk for the unborn should be preferred in women planning pregnancy. After inadvertent exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, close monitoring is necessary in pregnant women with supraventricular arrhythmias or cardiac disease.
Introduction
Ivabradine is a heart rate-lowering agent, acting through selective inhibition of the If (I-funny) channel without additional hemodynamic effects on cardiac function. The approved treatment indications in Europe are chronic heart failure and symptomatic treatment of chronic stable angina pectoris after failing or intolerance of beta-blockers [1].
Preclinical studies have suggested fetotoxicity and teratogenic effects. In pregnant rats, ivabradine exposure equivalent to therapeutic levels in humans was associated with increased intrauterine and postnatal mortality. Ventricular septal defects and complex anomalies of the great arteries were noted at doses 3 times the therapeutic human exposure. In pregnant rabbits, reduced fetal and placental weight as well as ectrodactyly was observed at doses 15 times the therapeutic human exposure [2]. In addition, dose-dependent mortality was seen in chicken and mice embryos [3, 4].
ESC guidelines consider ivabradine contraindicated during pregnancy and recommend discontinuation independent of treatment indication before conception with close clinical and echocardiographic monitoring [5].
Published data on ivabradine-exposed pregnancies are limited to three case reports, one with ivabradine treatment in the late first and early second trimester for sinus tachycardia after myocardial infarction [6], and the second starting ivabradine in the second trimester for deteriorating tachycardia-induced cardiomyopathy [7]. In the third case with first trimester exposure, pregnancy was electively terminated [8].
The limited human experience along with the teratogenic effects in animal studies makes it difficult to counsel inadvertently exposed pregnant women and their health care providers (HCP). Therefore, we decided to evaluate ivabradine-exposed pregnancies recorded by the German Embryotox Institute.
Methods
Study design
The German Embryotox Institute offers risk assessment on drug exposure during pregnancy to HCPs and patients. Up to 15,000 annual requests are answered, and in approximately 3500 critically exposed cases per year, the course and outcome of pregnancy are documented. At the initial contact, maternal characteristics and detailed exposure assessment are asked for after informed consent. About 8 weeks after the estimated date of birth, information on course and outcome of pregnancy is collected via structured telephone interview and mailed questionnaires. After a case by case plausibility check, additional information and health care records are requested in cases of adverse outcomes or inconsistent information.
Weeks of pregnancy were based on ultrasound and if not available on the last menstrual period (LMP). First trimester was defined as gestational week 2 + 0 (conception) until 12 + 6. Major birth defects were classified according to EUROCAT [9]. In prospectively ascertained pregnancies, neither the outcome of pregnancy nor prenatal pathology was known at the time of first contact. Pregnancies that were reported after birth or because of prenatal pathology were considered retrospective and evaluated separately. A detailed description of methodology adapted to the recommendations of the Strengthening of the Reporting of Observational studies in Epidemiology (STROBE) statement is given in Schaefer et al. [10] and Dathe and Schaefer [11].
We evaluated all requests on ivabradine to the German Embryotox Institute from 2007 to 2019.
Statistical analyses
Descriptive statistics were applied. For age, BMI, gestational weeks, neonatal weight, length, and head circumference, median and interquartile ranges are presented. Birth weights were adjusted to sex and gestational age at birth and percentile categories and standard deviation scores (SDS) were calculated according to the German perinatal survey [12].
Results
During the study period 2007–2019, our institute received 97 requests for ivabradine, 56 (58%) were from HCP, and 41 from patients. For further details, see Fig. 1. Treatment indications for ivabradine over time in relation to formal approval status and pregnancy labelling are summarized in Fig. 2. Supraventricular tachycardia was the most common reason for ivabradine treatment in our cohort. The first ivabradine-exposed pregnancy was recorded in March 2007, 1 year after approval and marketing authorization in Germany in January 2006.Fig. 1 Overview on information requests to the German Embryotox institute on ivabradine
Fig. 2 Treatment indication for ivabradine in 97 Embryotox requests over time in relation to pregnancy labelling and approval status by EMA/FDA. Recommendations for use of ivabradine in supraventricular tachycardias, i.e. inappropriate sinus tachycardia and postural orthostatic tachycardia syndrome: HRS/ESC 2015 [13], ACC/AHA/HFSA 2016 [14], ESC 2019 [15]. ACC, American College of Cardiology; EMA, European Medicines Agency; ESC, European Society of Cardiology; FDA, US Food and Drug Administration; HRS, Heart Rhythm Society
Fifty-six of the 97 requests were related to maternal exposure during pregnancy (Fig. 1). In 41 of these cases, follow-up on pregnancy outcome could be completed: 38 pregnancies were prospectively ascertained and three were retrospective reports.
Evaluation of prospectively ascertained pregnancies
Maternal characteristics
The median age of pregnant women was 29.5 years and the median BMI 22.5. Maternal characteristics are presented in detail in Table 1. The mean time at first contact was gestational week (GW) 7. Only 16 of the 38 prospectively ascertained pregnancies were actually planned (42%) and only 7 women (18%) started folic acid before conception. Contraceptive methods, mostly oral contraceptives (n = 4), were still used by 6 women at conception.Table 1 Maternal characteristics and pregnancy outcomes of ivabradine exposed pregnancies. Data are n (%) except for age, BMI, gestational week at birth, neonatal weight, length and head circumference, which are median and interquartile range
Maternal characteristics N (%) or median (IQR)
Maternal characteristics Age (n=38) 29.5 (26.5-34)
BMI (n=34) 22.5 (19.4-25.6)
Smoking (n=37)
>5 cigarettes/day 7 (19%)
<=5 cigarettes/day 4 (11%)
Alcohol (n=36)
<=1 drink/day 2 (6%)
>1 drink/day 1 (3%)
GW at first contact (n=38) 7 (5.6-11.1)
Previous pregnancies (n=38)
0 17(45%)
1 8 (21%)
2 13 (34%)
Pregnancy outcome and neonatal characteristics N (%) or median (IQR)
Pregnancy outcome (n=38) Spontaneous abortion 3
ETOP 3
Live birth 32
Pregnancy complications (n=32, only live births) Gestational diabetes 5 (16%)
Pre-eclampsia 2 (6%)
Cesarean section 15 (46%)
Preterm birth (<GW 37) 3 (9%)
Neonatal characteristics (n=33, including one pair of twins) GW at birth (n=32) 39 (37.9-39.9)
Weight, g (n=32) 3300 (2935-3610)
Length, cm (n=31) 50 (49-52.5)
HC, cm (n=30) 35 (34-36)
SGA (n=32) 2 (6%)
BMI, Body Mass Index; ETOP, elective termination of pregnancy; GW, gestational week; HC, head circumference; IQR, interquartile range; SGA, small for gestational age (<10th percentile)
Ivabradine exposure and co-medication
The most prevalent indication for ivabradine treatment was supraventricular tachycardia (n = 29) followed by heart failure (n = 3) and coronary artery disease (n = 2). Other reported indications (n = 4) included QT prolongation and sick-sinus-syndrome, although both are regarded as contraindications for ivabradine. The median daily ivabradine dose was 7.5 mg (IQR 5–10, min–max 2.5–15, n = 33).
Nearly all women were treated with ivabradine during the first trimester (n = 37) and 32 stopped treatment after recognition of pregnancy (Fig. 3), at median gestational week 6.1 (interquartile range 5.4–7.8). Cardiac decompensation or complications were not described after cessation of treatment. Only 5 women continued ivabradine throughout pregnancy or until shortly before delivery. One woman started treatment after first trimester (GW 18) of her twin pregnancy due to symptomatic tachycardia under beta-blocker therapy (Iv34 in Fig. 3).Fig. 3 Ivabradine exposure and pregnancy outcome of prospectively (n = 38) and retrospectively (n = 3) ascertained pregnancies
Cardiovascular co-medication consisted of beta-blockers (mainly metoprolol) in 15 women, mostly concomitantly (n = 7) or in replacement of ivabradine (n = 6). Two women initiated ivabradine after discontinuation of beta-blockers (Iv30 and Iv34 in Fig. 3). Six women received other cardiovascular medication, three methyldopa or amlodipine for hypertension, two digitoxin or verapamil for tachyarrhythmia, and one sildenafil and eplerenone for cardiomyopathy with cardiac failure (Iv17).
Psychiatric co-medication was reported in 7/38 women (5 of these with an anxiety disorder) including the teratogenic antiepileptic topiramate (n = 1) and pregabalin (n = 3).
Pregnancy outcome
Of the 38 prospectively ascertained ivabradine-exposed pregnancies, 32 were live births (33 infants including one pair of twins) and three resulted in early spontaneous abortions (Iv10, Iv17, and Iv19 in Fig. 3). One of these occurred in a woman with cardiomyopathy (Iv17) and various co-medications including hydroxycarbamide for thrombocythemia until GW 7. Three pregnancies were electively terminated for social/psychological reasons. Further information on pregnancy complications and outcomes is given in Table 1 and Fig. 3.
Major birth defects were reported in one pregnancy (Iv24) with 15 mg ivabradine plus bisoprolol until GW 7 + 5, both replaced by metoprolol. Polyhydramnios and preeclampsia led to Caesarean section in GW 35 after premature rupture of membranes. The premature child was large for gestational age (> 97 percentile) and had a cleft palate, an atrial septal defect II (spontaneously closed at the age of 10 months), and a hypertrophic cardiac septum. The mother was treated for sinus tachycardia, in addition she had diabetes type I and a BMI of 30.
Neonatal characteristics
Characteristics of the 33 neonates are summarized in Table 1. Compared to the German perinatal survey, standardized birth weights were lower with a median SDS of − 0.36, corresponding to a reduction of birth weight of 149 g for singletons born in gestational week 40.
Two neonates were small for gestational age, one of the twins (exposure only during 2nd and 3rd trimester, Iv34) and one after maternal ivabradine exposure until GW 12 (Iv29). Two neonates were large for gestational age, both mothers were diabetic (Iv24, type I diabetes and Iv27, gestational diabetes).
Retrospective cases
During the study period three pregnancies were retrospectively reported (further details are given in Table 2 and Fig. 3). One premature neonate (IvR1) with tracheal atresia died perinatally. The mother had been treated with ivabradine, naproxen, and metamizole during the first trimester until recognition of pregnancy (no further details available).Table 2 Retrospectively reported pregnancies
Case Maternal age/BMI Treatment indication Ivabradine exposure, other co-medication (GW or trimester) Pregnancy outcome, GW, birth weight percentile, sex Pregnancy complications, birth defects, and additional findings
IvR1 31 years/31.2 Tachycardia Ivabradine (1st trim)
Naproxen (1st trim)
Metamizole (1st trim)
Live birth, 36 + 2, > 97th, f Polyhydramnios, perinatal death, tracheal atresia, normal karyotype (46, XX)
IvR2 29 years/18.4 Tachycardia, syncopes, pacemaker Ivabradine (1st trim)
Bisoprolol (0–35 + 5)
Levetiracetam (0–35 + 5, grand mal epilepsy)
Live birth, 35 + 5, < 10th, f IUGR (since GW 30), oligohydramnios (since GW 34), Caesarean section (vaginal bleeding)
IvR3 29 years/not available Tachycardia Ivabradine (0–41 + 3)
Bisoprolol (0–41 + 3)
Live birth, 41 + 3
> 10th, m
Uncomplicated pregnancy
BMI, body mass index; f, female; GW, gestational week; IUGR, intrauterine growth restriction; m, male; trim, trimester
The other two retrospective cases include a woman with comorbidities, intrauterine growth restriction, and premature birth (IvR2). The second was a healthy newborn (IvR3) after treatment with ivabradine throughout pregnancy. This mother reported a previous also uneventful pregnancy course under ivabradine.
Discussion
Pregnancy outcome
In our prospective cohort, there was only one neonate with major birth defects (cleft palate and atrial septal defect) from a diabetic mother with supraventricular tachycardia. No further major or cardiac malformations were reported among the 32 live births and 6 pregnancy losses. Cardiac defects have been observed in preclinical experimental studies. As a possible teratogenic mechanism, ivabradine-induced embryonic heart rate reduction leading to reduced cardiac output and hypoxia was discussed [16]. With a prevalence of almost 1 of 100 cardiac malformations belong to the most frequent birth defects [17, 18]. In our case, the neonate was large for gestational age and had a septum hypertrophy, supporting the assumption that maternal pre-gestational diabetes may also be causative for the observed cardiac defect and cleft palate. A causal relationship between ivabradine and cleft palate is unlikely, because ivabradine was discontinued at GW 7 + 5 before the beginning of the critical time period for the development of cleft palate at gestational week 9. The rate of major birth defects in our prospective cohort is comparable to the expected background rate of 2–3% [17, 18]. However, the limited number of pregnancies exposed during the entire period of embryogenesis does not allow excluding an increased risk, so far.
There was only one retrospectively reported case with a birth defect (malformation of the trachea) suspected as potentially ivabradine related. The concomitant exposures with naproxen and metamizole have not been evidenced as teratogens in human, whereas the high maternal BMI in the present case poses a risk for birth defects.
Taken together, we did not observe a specific pattern of congenital anomalies in our study. In addition, the presence of maternal comorbidities like diabetes and high BMI may have contributed to the observed birth defects.
The number of spontaneous abortions (3/38) was not higher than expected. The rate of preeclampsia and preterm birth in our cohort was within the expected range [19].
Treatment indications
Although the approved treatment indications for ivabradine, chronic heart failure and chronic stable angina pectoris (see Fig. 2 for changes of licenced use during the study period), are rare in women of reproductive age, we have observed an increasing need for information for off-label use in (younger) women considering pregnancy.
In our case series, most women were treated (off-label) for supraventricular tachycardia using lower doses than recommended (5–7.5 mg twice daily). Supraventricular tachycardia is not uncommon, especially in younger women, and is expected in 0.02–0.5% of pregnancies [20]. Aggravation of symptoms during pregnancy is described in about 20–50% of women related to various mechanisms, including pregnancy-related increase in cardiac heart rate and output and hormonal changes [20].
In 2015, the HRS (Heart Rhythm Society) considered ivabradine as promising for treating patients with inappropriate sinus tachycardia (IST) [13], followed by a recommendation for ivabradine by the ACC (American College of Cardiology) in 2016 [14]—though only one small randomized controlled study has yet supported beneficial effects [21]. Although still not approved for this indication, recent ESC guidelines [15] also recommend ivabradine alone or in combination with a beta-blocker in symptomatic patients with IST, for the treatment of focal atrial tachycardia or for postural orthostatic tachycardia syndrome (POTS), a condition considered as common cause for orthostatic intolerance in women between 15 and 25 years [20, 22] (Fig. 2).
Beta-blockers were required in 18% of women in our prospective cohort (Fig. 3). Discontinuation of ivabradine or switching to beta-blockers after recognition of pregnancy was well tolerated in most of our patients. Thus, necessity of ivabradine treatment should be critically considered in women with supraventricular tachycardia planning pregnancy. Data from previous studies suggest that discontinuation of ivabradine is possible in many (non-pregnant) patients with IST without recurrence. Even beneficial long-term effects after cessation of therapy are discussed [23, 24].
A high number of women in our cohort received therapy for anxiety and/or other psychiatric conditions. This is in line with findings from other studies reporting high rates of anxiety (25%) and depression (26%) in IST patients [25]. Anxiety is considered as non-cardiac symptom of inappropriate sinus tachycardia, and vice versa, anxiety may trigger tachycardia. Especially during pregnancy, empathic care and reassurance are necessary, to prevent increased concerns about possible harm to the fetus [26].
Preconception counselling is of utmost importance in all women considering pregnancy under medication [27]. Sinus tachycardia is a condition diagnosed mainly in young females and the first diagnosis is frequently made during or after pregnancy [25]. Considering the increasing off-label use of ivabradine as well as the high rate of > 40% unplanned pregnancies [28], more exposed pregnancies can be expected in the future. Well-established treatment options such as selected beta-blockers with low reproductive risk [29] should be considered in women of childbearing age before prescription of ivabradine.
Strength and limitations
A major strength of our study is the detailed exposure protocol and the multi-source ascertainment of pregnancy outcome via patients and their HCP followed by a case by case plausibility test.
Our data may not be representative of the German pregnant population; insofar, women with higher education are over-represented [30]. However, the regional distribution of enrolled patients is representative of the female population of childbearing age in Germany [11]. In addition, women on long-term ivabradine with severe cardiovascular conditions such as the approved treatment indications coronary artery disease and cardiac failure are under-represented in our cohort possibly favouring uneventful pregnancy outcomes. However, given the rarity of severe cardiac conditions in pregnant women, large study cohorts are not to be expected in the foreseeable future.
Conclusions
Our findings do not indicate that ivabradine is a major teratogen when used in early pregnancy. Although being the largest study published so far on the outcome of ivabradine-exposed pregnancies, the sample is too small to rule out embryotoxic effects. Further studies are needed to confirm or refute our findings.
In women planning pregnancy, established drugs of choice with low risk for the unborn should be preferred to ivabradine. For women with significant arrhythmias and severe cardiac disease, the impact of the underlying medical condition including comorbidities such as diabetes and high BMI on pregnancy outcome has to be considered when planning pregnancy.
After inadvertent ivabradine exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, all pregnant women with complicated arrhythmias or major cardiac disease should be carefully monitored.
Acknowledgements
We would like to thank all patients and physicians reporting pregnancies to the institute and acknowledge our colleagues from the German Embryotox Institute for counselling and thoroughly documenting cases, especially Sandra Zinke for continuous support with case documentation.
Contributions of authors
MH, ML, and CS designed the study. ML, EB, MH, and KD validated and analysed the data. ML, MH, and CS wrote the first draft of the manuscript, and all the authors critically revised subsequent manuscript drafts and contributed essential discussion points. All the authors approved the final version of this manuscript and they are responsible for the accuracy of this work.
Funding
Open Access funding enabled and organized by Projekt DEAL. This work was performed with financial support from the German Ministry of Health (BMG) and the German Federal Institute for Drugs and Medical Devices (BfArM). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
The study was registered with the German Clinical Trial register (DRKS00010502) and the study protocol was approved by the ethics committee of the Charité-Universitätsmedizin Berlin (EA1/107/16).
The study was registered with the German Clinical Trial register (DRKS00010502, 20.5.2016)
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | BISOPROLOL, IVABRADINE, METOPROLOL | DrugsGivenReaction | CC BY | 33501507 | 19,697,614 | 2021-07 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Cleft palate'. | Ivabradine use in pregnant women-treatment indications and pregnancy outcome: an evaluation of the German Embryotox database.
OBJECTIVE
Ivabradine has been approved for the treatment of chronic heart failure and chronic stable angina pectoris in Europe. Based on adverse outcomes of reproductive animal studies and the lack of human data, ivabradine is considered contraindicated during pregnancy. The aim of this observational study is to analyse ivabradine use before and during pregnancy.
METHODS
We evaluated all ivabradine-related requests to the German Embryotox Institute from 2007 to 2019. Exposed pregnancies were analysed as to their outcome.
RESULTS
Off-label use for supraventricular tachycardia was frequent in women of childbearing age. Of 38 prospectively ascertained pregnancies with ivabradine exposure and completed follow-up, 32 resulted in live births, 3 in spontaneous abortions, and 3 were electively terminated. One neonate presented with major birth defects (atrial septal defect and cleft palate). In 33/38 patients, ivabradine was discontinued after confirmation of pregnancy without cardiac deterioration and 5/38 women continued ivabradine throughout pregnancy. In addition, there were 3 retrospectively reported pregnancies including one major birth defect (tracheal atresia).
CONCLUSIONS
This case series represents the largest cohort of ivabradine-exposed pregnancies, published so far. According to our findings, ivabradine appears not to be a major teratogen. However, established drugs of choice with strong evidence of low risk for the unborn should be preferred in women planning pregnancy. After inadvertent exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, close monitoring is necessary in pregnant women with supraventricular arrhythmias or cardiac disease.
Introduction
Ivabradine is a heart rate-lowering agent, acting through selective inhibition of the If (I-funny) channel without additional hemodynamic effects on cardiac function. The approved treatment indications in Europe are chronic heart failure and symptomatic treatment of chronic stable angina pectoris after failing or intolerance of beta-blockers [1].
Preclinical studies have suggested fetotoxicity and teratogenic effects. In pregnant rats, ivabradine exposure equivalent to therapeutic levels in humans was associated with increased intrauterine and postnatal mortality. Ventricular septal defects and complex anomalies of the great arteries were noted at doses 3 times the therapeutic human exposure. In pregnant rabbits, reduced fetal and placental weight as well as ectrodactyly was observed at doses 15 times the therapeutic human exposure [2]. In addition, dose-dependent mortality was seen in chicken and mice embryos [3, 4].
ESC guidelines consider ivabradine contraindicated during pregnancy and recommend discontinuation independent of treatment indication before conception with close clinical and echocardiographic monitoring [5].
Published data on ivabradine-exposed pregnancies are limited to three case reports, one with ivabradine treatment in the late first and early second trimester for sinus tachycardia after myocardial infarction [6], and the second starting ivabradine in the second trimester for deteriorating tachycardia-induced cardiomyopathy [7]. In the third case with first trimester exposure, pregnancy was electively terminated [8].
The limited human experience along with the teratogenic effects in animal studies makes it difficult to counsel inadvertently exposed pregnant women and their health care providers (HCP). Therefore, we decided to evaluate ivabradine-exposed pregnancies recorded by the German Embryotox Institute.
Methods
Study design
The German Embryotox Institute offers risk assessment on drug exposure during pregnancy to HCPs and patients. Up to 15,000 annual requests are answered, and in approximately 3500 critically exposed cases per year, the course and outcome of pregnancy are documented. At the initial contact, maternal characteristics and detailed exposure assessment are asked for after informed consent. About 8 weeks after the estimated date of birth, information on course and outcome of pregnancy is collected via structured telephone interview and mailed questionnaires. After a case by case plausibility check, additional information and health care records are requested in cases of adverse outcomes or inconsistent information.
Weeks of pregnancy were based on ultrasound and if not available on the last menstrual period (LMP). First trimester was defined as gestational week 2 + 0 (conception) until 12 + 6. Major birth defects were classified according to EUROCAT [9]. In prospectively ascertained pregnancies, neither the outcome of pregnancy nor prenatal pathology was known at the time of first contact. Pregnancies that were reported after birth or because of prenatal pathology were considered retrospective and evaluated separately. A detailed description of methodology adapted to the recommendations of the Strengthening of the Reporting of Observational studies in Epidemiology (STROBE) statement is given in Schaefer et al. [10] and Dathe and Schaefer [11].
We evaluated all requests on ivabradine to the German Embryotox Institute from 2007 to 2019.
Statistical analyses
Descriptive statistics were applied. For age, BMI, gestational weeks, neonatal weight, length, and head circumference, median and interquartile ranges are presented. Birth weights were adjusted to sex and gestational age at birth and percentile categories and standard deviation scores (SDS) were calculated according to the German perinatal survey [12].
Results
During the study period 2007–2019, our institute received 97 requests for ivabradine, 56 (58%) were from HCP, and 41 from patients. For further details, see Fig. 1. Treatment indications for ivabradine over time in relation to formal approval status and pregnancy labelling are summarized in Fig. 2. Supraventricular tachycardia was the most common reason for ivabradine treatment in our cohort. The first ivabradine-exposed pregnancy was recorded in March 2007, 1 year after approval and marketing authorization in Germany in January 2006.Fig. 1 Overview on information requests to the German Embryotox institute on ivabradine
Fig. 2 Treatment indication for ivabradine in 97 Embryotox requests over time in relation to pregnancy labelling and approval status by EMA/FDA. Recommendations for use of ivabradine in supraventricular tachycardias, i.e. inappropriate sinus tachycardia and postural orthostatic tachycardia syndrome: HRS/ESC 2015 [13], ACC/AHA/HFSA 2016 [14], ESC 2019 [15]. ACC, American College of Cardiology; EMA, European Medicines Agency; ESC, European Society of Cardiology; FDA, US Food and Drug Administration; HRS, Heart Rhythm Society
Fifty-six of the 97 requests were related to maternal exposure during pregnancy (Fig. 1). In 41 of these cases, follow-up on pregnancy outcome could be completed: 38 pregnancies were prospectively ascertained and three were retrospective reports.
Evaluation of prospectively ascertained pregnancies
Maternal characteristics
The median age of pregnant women was 29.5 years and the median BMI 22.5. Maternal characteristics are presented in detail in Table 1. The mean time at first contact was gestational week (GW) 7. Only 16 of the 38 prospectively ascertained pregnancies were actually planned (42%) and only 7 women (18%) started folic acid before conception. Contraceptive methods, mostly oral contraceptives (n = 4), were still used by 6 women at conception.Table 1 Maternal characteristics and pregnancy outcomes of ivabradine exposed pregnancies. Data are n (%) except for age, BMI, gestational week at birth, neonatal weight, length and head circumference, which are median and interquartile range
Maternal characteristics N (%) or median (IQR)
Maternal characteristics Age (n=38) 29.5 (26.5-34)
BMI (n=34) 22.5 (19.4-25.6)
Smoking (n=37)
>5 cigarettes/day 7 (19%)
<=5 cigarettes/day 4 (11%)
Alcohol (n=36)
<=1 drink/day 2 (6%)
>1 drink/day 1 (3%)
GW at first contact (n=38) 7 (5.6-11.1)
Previous pregnancies (n=38)
0 17(45%)
1 8 (21%)
2 13 (34%)
Pregnancy outcome and neonatal characteristics N (%) or median (IQR)
Pregnancy outcome (n=38) Spontaneous abortion 3
ETOP 3
Live birth 32
Pregnancy complications (n=32, only live births) Gestational diabetes 5 (16%)
Pre-eclampsia 2 (6%)
Cesarean section 15 (46%)
Preterm birth (<GW 37) 3 (9%)
Neonatal characteristics (n=33, including one pair of twins) GW at birth (n=32) 39 (37.9-39.9)
Weight, g (n=32) 3300 (2935-3610)
Length, cm (n=31) 50 (49-52.5)
HC, cm (n=30) 35 (34-36)
SGA (n=32) 2 (6%)
BMI, Body Mass Index; ETOP, elective termination of pregnancy; GW, gestational week; HC, head circumference; IQR, interquartile range; SGA, small for gestational age (<10th percentile)
Ivabradine exposure and co-medication
The most prevalent indication for ivabradine treatment was supraventricular tachycardia (n = 29) followed by heart failure (n = 3) and coronary artery disease (n = 2). Other reported indications (n = 4) included QT prolongation and sick-sinus-syndrome, although both are regarded as contraindications for ivabradine. The median daily ivabradine dose was 7.5 mg (IQR 5–10, min–max 2.5–15, n = 33).
Nearly all women were treated with ivabradine during the first trimester (n = 37) and 32 stopped treatment after recognition of pregnancy (Fig. 3), at median gestational week 6.1 (interquartile range 5.4–7.8). Cardiac decompensation or complications were not described after cessation of treatment. Only 5 women continued ivabradine throughout pregnancy or until shortly before delivery. One woman started treatment after first trimester (GW 18) of her twin pregnancy due to symptomatic tachycardia under beta-blocker therapy (Iv34 in Fig. 3).Fig. 3 Ivabradine exposure and pregnancy outcome of prospectively (n = 38) and retrospectively (n = 3) ascertained pregnancies
Cardiovascular co-medication consisted of beta-blockers (mainly metoprolol) in 15 women, mostly concomitantly (n = 7) or in replacement of ivabradine (n = 6). Two women initiated ivabradine after discontinuation of beta-blockers (Iv30 and Iv34 in Fig. 3). Six women received other cardiovascular medication, three methyldopa or amlodipine for hypertension, two digitoxin or verapamil for tachyarrhythmia, and one sildenafil and eplerenone for cardiomyopathy with cardiac failure (Iv17).
Psychiatric co-medication was reported in 7/38 women (5 of these with an anxiety disorder) including the teratogenic antiepileptic topiramate (n = 1) and pregabalin (n = 3).
Pregnancy outcome
Of the 38 prospectively ascertained ivabradine-exposed pregnancies, 32 were live births (33 infants including one pair of twins) and three resulted in early spontaneous abortions (Iv10, Iv17, and Iv19 in Fig. 3). One of these occurred in a woman with cardiomyopathy (Iv17) and various co-medications including hydroxycarbamide for thrombocythemia until GW 7. Three pregnancies were electively terminated for social/psychological reasons. Further information on pregnancy complications and outcomes is given in Table 1 and Fig. 3.
Major birth defects were reported in one pregnancy (Iv24) with 15 mg ivabradine plus bisoprolol until GW 7 + 5, both replaced by metoprolol. Polyhydramnios and preeclampsia led to Caesarean section in GW 35 after premature rupture of membranes. The premature child was large for gestational age (> 97 percentile) and had a cleft palate, an atrial septal defect II (spontaneously closed at the age of 10 months), and a hypertrophic cardiac septum. The mother was treated for sinus tachycardia, in addition she had diabetes type I and a BMI of 30.
Neonatal characteristics
Characteristics of the 33 neonates are summarized in Table 1. Compared to the German perinatal survey, standardized birth weights were lower with a median SDS of − 0.36, corresponding to a reduction of birth weight of 149 g for singletons born in gestational week 40.
Two neonates were small for gestational age, one of the twins (exposure only during 2nd and 3rd trimester, Iv34) and one after maternal ivabradine exposure until GW 12 (Iv29). Two neonates were large for gestational age, both mothers were diabetic (Iv24, type I diabetes and Iv27, gestational diabetes).
Retrospective cases
During the study period three pregnancies were retrospectively reported (further details are given in Table 2 and Fig. 3). One premature neonate (IvR1) with tracheal atresia died perinatally. The mother had been treated with ivabradine, naproxen, and metamizole during the first trimester until recognition of pregnancy (no further details available).Table 2 Retrospectively reported pregnancies
Case Maternal age/BMI Treatment indication Ivabradine exposure, other co-medication (GW or trimester) Pregnancy outcome, GW, birth weight percentile, sex Pregnancy complications, birth defects, and additional findings
IvR1 31 years/31.2 Tachycardia Ivabradine (1st trim)
Naproxen (1st trim)
Metamizole (1st trim)
Live birth, 36 + 2, > 97th, f Polyhydramnios, perinatal death, tracheal atresia, normal karyotype (46, XX)
IvR2 29 years/18.4 Tachycardia, syncopes, pacemaker Ivabradine (1st trim)
Bisoprolol (0–35 + 5)
Levetiracetam (0–35 + 5, grand mal epilepsy)
Live birth, 35 + 5, < 10th, f IUGR (since GW 30), oligohydramnios (since GW 34), Caesarean section (vaginal bleeding)
IvR3 29 years/not available Tachycardia Ivabradine (0–41 + 3)
Bisoprolol (0–41 + 3)
Live birth, 41 + 3
> 10th, m
Uncomplicated pregnancy
BMI, body mass index; f, female; GW, gestational week; IUGR, intrauterine growth restriction; m, male; trim, trimester
The other two retrospective cases include a woman with comorbidities, intrauterine growth restriction, and premature birth (IvR2). The second was a healthy newborn (IvR3) after treatment with ivabradine throughout pregnancy. This mother reported a previous also uneventful pregnancy course under ivabradine.
Discussion
Pregnancy outcome
In our prospective cohort, there was only one neonate with major birth defects (cleft palate and atrial septal defect) from a diabetic mother with supraventricular tachycardia. No further major or cardiac malformations were reported among the 32 live births and 6 pregnancy losses. Cardiac defects have been observed in preclinical experimental studies. As a possible teratogenic mechanism, ivabradine-induced embryonic heart rate reduction leading to reduced cardiac output and hypoxia was discussed [16]. With a prevalence of almost 1 of 100 cardiac malformations belong to the most frequent birth defects [17, 18]. In our case, the neonate was large for gestational age and had a septum hypertrophy, supporting the assumption that maternal pre-gestational diabetes may also be causative for the observed cardiac defect and cleft palate. A causal relationship between ivabradine and cleft palate is unlikely, because ivabradine was discontinued at GW 7 + 5 before the beginning of the critical time period for the development of cleft palate at gestational week 9. The rate of major birth defects in our prospective cohort is comparable to the expected background rate of 2–3% [17, 18]. However, the limited number of pregnancies exposed during the entire period of embryogenesis does not allow excluding an increased risk, so far.
There was only one retrospectively reported case with a birth defect (malformation of the trachea) suspected as potentially ivabradine related. The concomitant exposures with naproxen and metamizole have not been evidenced as teratogens in human, whereas the high maternal BMI in the present case poses a risk for birth defects.
Taken together, we did not observe a specific pattern of congenital anomalies in our study. In addition, the presence of maternal comorbidities like diabetes and high BMI may have contributed to the observed birth defects.
The number of spontaneous abortions (3/38) was not higher than expected. The rate of preeclampsia and preterm birth in our cohort was within the expected range [19].
Treatment indications
Although the approved treatment indications for ivabradine, chronic heart failure and chronic stable angina pectoris (see Fig. 2 for changes of licenced use during the study period), are rare in women of reproductive age, we have observed an increasing need for information for off-label use in (younger) women considering pregnancy.
In our case series, most women were treated (off-label) for supraventricular tachycardia using lower doses than recommended (5–7.5 mg twice daily). Supraventricular tachycardia is not uncommon, especially in younger women, and is expected in 0.02–0.5% of pregnancies [20]. Aggravation of symptoms during pregnancy is described in about 20–50% of women related to various mechanisms, including pregnancy-related increase in cardiac heart rate and output and hormonal changes [20].
In 2015, the HRS (Heart Rhythm Society) considered ivabradine as promising for treating patients with inappropriate sinus tachycardia (IST) [13], followed by a recommendation for ivabradine by the ACC (American College of Cardiology) in 2016 [14]—though only one small randomized controlled study has yet supported beneficial effects [21]. Although still not approved for this indication, recent ESC guidelines [15] also recommend ivabradine alone or in combination with a beta-blocker in symptomatic patients with IST, for the treatment of focal atrial tachycardia or for postural orthostatic tachycardia syndrome (POTS), a condition considered as common cause for orthostatic intolerance in women between 15 and 25 years [20, 22] (Fig. 2).
Beta-blockers were required in 18% of women in our prospective cohort (Fig. 3). Discontinuation of ivabradine or switching to beta-blockers after recognition of pregnancy was well tolerated in most of our patients. Thus, necessity of ivabradine treatment should be critically considered in women with supraventricular tachycardia planning pregnancy. Data from previous studies suggest that discontinuation of ivabradine is possible in many (non-pregnant) patients with IST without recurrence. Even beneficial long-term effects after cessation of therapy are discussed [23, 24].
A high number of women in our cohort received therapy for anxiety and/or other psychiatric conditions. This is in line with findings from other studies reporting high rates of anxiety (25%) and depression (26%) in IST patients [25]. Anxiety is considered as non-cardiac symptom of inappropriate sinus tachycardia, and vice versa, anxiety may trigger tachycardia. Especially during pregnancy, empathic care and reassurance are necessary, to prevent increased concerns about possible harm to the fetus [26].
Preconception counselling is of utmost importance in all women considering pregnancy under medication [27]. Sinus tachycardia is a condition diagnosed mainly in young females and the first diagnosis is frequently made during or after pregnancy [25]. Considering the increasing off-label use of ivabradine as well as the high rate of > 40% unplanned pregnancies [28], more exposed pregnancies can be expected in the future. Well-established treatment options such as selected beta-blockers with low reproductive risk [29] should be considered in women of childbearing age before prescription of ivabradine.
Strength and limitations
A major strength of our study is the detailed exposure protocol and the multi-source ascertainment of pregnancy outcome via patients and their HCP followed by a case by case plausibility test.
Our data may not be representative of the German pregnant population; insofar, women with higher education are over-represented [30]. However, the regional distribution of enrolled patients is representative of the female population of childbearing age in Germany [11]. In addition, women on long-term ivabradine with severe cardiovascular conditions such as the approved treatment indications coronary artery disease and cardiac failure are under-represented in our cohort possibly favouring uneventful pregnancy outcomes. However, given the rarity of severe cardiac conditions in pregnant women, large study cohorts are not to be expected in the foreseeable future.
Conclusions
Our findings do not indicate that ivabradine is a major teratogen when used in early pregnancy. Although being the largest study published so far on the outcome of ivabradine-exposed pregnancies, the sample is too small to rule out embryotoxic effects. Further studies are needed to confirm or refute our findings.
In women planning pregnancy, established drugs of choice with low risk for the unborn should be preferred to ivabradine. For women with significant arrhythmias and severe cardiac disease, the impact of the underlying medical condition including comorbidities such as diabetes and high BMI on pregnancy outcome has to be considered when planning pregnancy.
After inadvertent ivabradine exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, all pregnant women with complicated arrhythmias or major cardiac disease should be carefully monitored.
Acknowledgements
We would like to thank all patients and physicians reporting pregnancies to the institute and acknowledge our colleagues from the German Embryotox Institute for counselling and thoroughly documenting cases, especially Sandra Zinke for continuous support with case documentation.
Contributions of authors
MH, ML, and CS designed the study. ML, EB, MH, and KD validated and analysed the data. ML, MH, and CS wrote the first draft of the manuscript, and all the authors critically revised subsequent manuscript drafts and contributed essential discussion points. All the authors approved the final version of this manuscript and they are responsible for the accuracy of this work.
Funding
Open Access funding enabled and organized by Projekt DEAL. This work was performed with financial support from the German Ministry of Health (BMG) and the German Federal Institute for Drugs and Medical Devices (BfArM). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
The study was registered with the German Clinical Trial register (DRKS00010502) and the study protocol was approved by the ethics committee of the Charité-Universitätsmedizin Berlin (EA1/107/16).
The study was registered with the German Clinical Trial register (DRKS00010502, 20.5.2016)
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | BISOPROLOL, IVABRADINE, METOPROLOL | DrugsGivenReaction | CC BY | 33501507 | 19,697,614 | 2021-07 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Exposure during pregnancy'. | Ivabradine use in pregnant women-treatment indications and pregnancy outcome: an evaluation of the German Embryotox database.
OBJECTIVE
Ivabradine has been approved for the treatment of chronic heart failure and chronic stable angina pectoris in Europe. Based on adverse outcomes of reproductive animal studies and the lack of human data, ivabradine is considered contraindicated during pregnancy. The aim of this observational study is to analyse ivabradine use before and during pregnancy.
METHODS
We evaluated all ivabradine-related requests to the German Embryotox Institute from 2007 to 2019. Exposed pregnancies were analysed as to their outcome.
RESULTS
Off-label use for supraventricular tachycardia was frequent in women of childbearing age. Of 38 prospectively ascertained pregnancies with ivabradine exposure and completed follow-up, 32 resulted in live births, 3 in spontaneous abortions, and 3 were electively terminated. One neonate presented with major birth defects (atrial septal defect and cleft palate). In 33/38 patients, ivabradine was discontinued after confirmation of pregnancy without cardiac deterioration and 5/38 women continued ivabradine throughout pregnancy. In addition, there were 3 retrospectively reported pregnancies including one major birth defect (tracheal atresia).
CONCLUSIONS
This case series represents the largest cohort of ivabradine-exposed pregnancies, published so far. According to our findings, ivabradine appears not to be a major teratogen. However, established drugs of choice with strong evidence of low risk for the unborn should be preferred in women planning pregnancy. After inadvertent exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, close monitoring is necessary in pregnant women with supraventricular arrhythmias or cardiac disease.
Introduction
Ivabradine is a heart rate-lowering agent, acting through selective inhibition of the If (I-funny) channel without additional hemodynamic effects on cardiac function. The approved treatment indications in Europe are chronic heart failure and symptomatic treatment of chronic stable angina pectoris after failing or intolerance of beta-blockers [1].
Preclinical studies have suggested fetotoxicity and teratogenic effects. In pregnant rats, ivabradine exposure equivalent to therapeutic levels in humans was associated with increased intrauterine and postnatal mortality. Ventricular septal defects and complex anomalies of the great arteries were noted at doses 3 times the therapeutic human exposure. In pregnant rabbits, reduced fetal and placental weight as well as ectrodactyly was observed at doses 15 times the therapeutic human exposure [2]. In addition, dose-dependent mortality was seen in chicken and mice embryos [3, 4].
ESC guidelines consider ivabradine contraindicated during pregnancy and recommend discontinuation independent of treatment indication before conception with close clinical and echocardiographic monitoring [5].
Published data on ivabradine-exposed pregnancies are limited to three case reports, one with ivabradine treatment in the late first and early second trimester for sinus tachycardia after myocardial infarction [6], and the second starting ivabradine in the second trimester for deteriorating tachycardia-induced cardiomyopathy [7]. In the third case with first trimester exposure, pregnancy was electively terminated [8].
The limited human experience along with the teratogenic effects in animal studies makes it difficult to counsel inadvertently exposed pregnant women and their health care providers (HCP). Therefore, we decided to evaluate ivabradine-exposed pregnancies recorded by the German Embryotox Institute.
Methods
Study design
The German Embryotox Institute offers risk assessment on drug exposure during pregnancy to HCPs and patients. Up to 15,000 annual requests are answered, and in approximately 3500 critically exposed cases per year, the course and outcome of pregnancy are documented. At the initial contact, maternal characteristics and detailed exposure assessment are asked for after informed consent. About 8 weeks after the estimated date of birth, information on course and outcome of pregnancy is collected via structured telephone interview and mailed questionnaires. After a case by case plausibility check, additional information and health care records are requested in cases of adverse outcomes or inconsistent information.
Weeks of pregnancy were based on ultrasound and if not available on the last menstrual period (LMP). First trimester was defined as gestational week 2 + 0 (conception) until 12 + 6. Major birth defects were classified according to EUROCAT [9]. In prospectively ascertained pregnancies, neither the outcome of pregnancy nor prenatal pathology was known at the time of first contact. Pregnancies that were reported after birth or because of prenatal pathology were considered retrospective and evaluated separately. A detailed description of methodology adapted to the recommendations of the Strengthening of the Reporting of Observational studies in Epidemiology (STROBE) statement is given in Schaefer et al. [10] and Dathe and Schaefer [11].
We evaluated all requests on ivabradine to the German Embryotox Institute from 2007 to 2019.
Statistical analyses
Descriptive statistics were applied. For age, BMI, gestational weeks, neonatal weight, length, and head circumference, median and interquartile ranges are presented. Birth weights were adjusted to sex and gestational age at birth and percentile categories and standard deviation scores (SDS) were calculated according to the German perinatal survey [12].
Results
During the study period 2007–2019, our institute received 97 requests for ivabradine, 56 (58%) were from HCP, and 41 from patients. For further details, see Fig. 1. Treatment indications for ivabradine over time in relation to formal approval status and pregnancy labelling are summarized in Fig. 2. Supraventricular tachycardia was the most common reason for ivabradine treatment in our cohort. The first ivabradine-exposed pregnancy was recorded in March 2007, 1 year after approval and marketing authorization in Germany in January 2006.Fig. 1 Overview on information requests to the German Embryotox institute on ivabradine
Fig. 2 Treatment indication for ivabradine in 97 Embryotox requests over time in relation to pregnancy labelling and approval status by EMA/FDA. Recommendations for use of ivabradine in supraventricular tachycardias, i.e. inappropriate sinus tachycardia and postural orthostatic tachycardia syndrome: HRS/ESC 2015 [13], ACC/AHA/HFSA 2016 [14], ESC 2019 [15]. ACC, American College of Cardiology; EMA, European Medicines Agency; ESC, European Society of Cardiology; FDA, US Food and Drug Administration; HRS, Heart Rhythm Society
Fifty-six of the 97 requests were related to maternal exposure during pregnancy (Fig. 1). In 41 of these cases, follow-up on pregnancy outcome could be completed: 38 pregnancies were prospectively ascertained and three were retrospective reports.
Evaluation of prospectively ascertained pregnancies
Maternal characteristics
The median age of pregnant women was 29.5 years and the median BMI 22.5. Maternal characteristics are presented in detail in Table 1. The mean time at first contact was gestational week (GW) 7. Only 16 of the 38 prospectively ascertained pregnancies were actually planned (42%) and only 7 women (18%) started folic acid before conception. Contraceptive methods, mostly oral contraceptives (n = 4), were still used by 6 women at conception.Table 1 Maternal characteristics and pregnancy outcomes of ivabradine exposed pregnancies. Data are n (%) except for age, BMI, gestational week at birth, neonatal weight, length and head circumference, which are median and interquartile range
Maternal characteristics N (%) or median (IQR)
Maternal characteristics Age (n=38) 29.5 (26.5-34)
BMI (n=34) 22.5 (19.4-25.6)
Smoking (n=37)
>5 cigarettes/day 7 (19%)
<=5 cigarettes/day 4 (11%)
Alcohol (n=36)
<=1 drink/day 2 (6%)
>1 drink/day 1 (3%)
GW at first contact (n=38) 7 (5.6-11.1)
Previous pregnancies (n=38)
0 17(45%)
1 8 (21%)
2 13 (34%)
Pregnancy outcome and neonatal characteristics N (%) or median (IQR)
Pregnancy outcome (n=38) Spontaneous abortion 3
ETOP 3
Live birth 32
Pregnancy complications (n=32, only live births) Gestational diabetes 5 (16%)
Pre-eclampsia 2 (6%)
Cesarean section 15 (46%)
Preterm birth (<GW 37) 3 (9%)
Neonatal characteristics (n=33, including one pair of twins) GW at birth (n=32) 39 (37.9-39.9)
Weight, g (n=32) 3300 (2935-3610)
Length, cm (n=31) 50 (49-52.5)
HC, cm (n=30) 35 (34-36)
SGA (n=32) 2 (6%)
BMI, Body Mass Index; ETOP, elective termination of pregnancy; GW, gestational week; HC, head circumference; IQR, interquartile range; SGA, small for gestational age (<10th percentile)
Ivabradine exposure and co-medication
The most prevalent indication for ivabradine treatment was supraventricular tachycardia (n = 29) followed by heart failure (n = 3) and coronary artery disease (n = 2). Other reported indications (n = 4) included QT prolongation and sick-sinus-syndrome, although both are regarded as contraindications for ivabradine. The median daily ivabradine dose was 7.5 mg (IQR 5–10, min–max 2.5–15, n = 33).
Nearly all women were treated with ivabradine during the first trimester (n = 37) and 32 stopped treatment after recognition of pregnancy (Fig. 3), at median gestational week 6.1 (interquartile range 5.4–7.8). Cardiac decompensation or complications were not described after cessation of treatment. Only 5 women continued ivabradine throughout pregnancy or until shortly before delivery. One woman started treatment after first trimester (GW 18) of her twin pregnancy due to symptomatic tachycardia under beta-blocker therapy (Iv34 in Fig. 3).Fig. 3 Ivabradine exposure and pregnancy outcome of prospectively (n = 38) and retrospectively (n = 3) ascertained pregnancies
Cardiovascular co-medication consisted of beta-blockers (mainly metoprolol) in 15 women, mostly concomitantly (n = 7) or in replacement of ivabradine (n = 6). Two women initiated ivabradine after discontinuation of beta-blockers (Iv30 and Iv34 in Fig. 3). Six women received other cardiovascular medication, three methyldopa or amlodipine for hypertension, two digitoxin or verapamil for tachyarrhythmia, and one sildenafil and eplerenone for cardiomyopathy with cardiac failure (Iv17).
Psychiatric co-medication was reported in 7/38 women (5 of these with an anxiety disorder) including the teratogenic antiepileptic topiramate (n = 1) and pregabalin (n = 3).
Pregnancy outcome
Of the 38 prospectively ascertained ivabradine-exposed pregnancies, 32 were live births (33 infants including one pair of twins) and three resulted in early spontaneous abortions (Iv10, Iv17, and Iv19 in Fig. 3). One of these occurred in a woman with cardiomyopathy (Iv17) and various co-medications including hydroxycarbamide for thrombocythemia until GW 7. Three pregnancies were electively terminated for social/psychological reasons. Further information on pregnancy complications and outcomes is given in Table 1 and Fig. 3.
Major birth defects were reported in one pregnancy (Iv24) with 15 mg ivabradine plus bisoprolol until GW 7 + 5, both replaced by metoprolol. Polyhydramnios and preeclampsia led to Caesarean section in GW 35 after premature rupture of membranes. The premature child was large for gestational age (> 97 percentile) and had a cleft palate, an atrial septal defect II (spontaneously closed at the age of 10 months), and a hypertrophic cardiac septum. The mother was treated for sinus tachycardia, in addition she had diabetes type I and a BMI of 30.
Neonatal characteristics
Characteristics of the 33 neonates are summarized in Table 1. Compared to the German perinatal survey, standardized birth weights were lower with a median SDS of − 0.36, corresponding to a reduction of birth weight of 149 g for singletons born in gestational week 40.
Two neonates were small for gestational age, one of the twins (exposure only during 2nd and 3rd trimester, Iv34) and one after maternal ivabradine exposure until GW 12 (Iv29). Two neonates were large for gestational age, both mothers were diabetic (Iv24, type I diabetes and Iv27, gestational diabetes).
Retrospective cases
During the study period three pregnancies were retrospectively reported (further details are given in Table 2 and Fig. 3). One premature neonate (IvR1) with tracheal atresia died perinatally. The mother had been treated with ivabradine, naproxen, and metamizole during the first trimester until recognition of pregnancy (no further details available).Table 2 Retrospectively reported pregnancies
Case Maternal age/BMI Treatment indication Ivabradine exposure, other co-medication (GW or trimester) Pregnancy outcome, GW, birth weight percentile, sex Pregnancy complications, birth defects, and additional findings
IvR1 31 years/31.2 Tachycardia Ivabradine (1st trim)
Naproxen (1st trim)
Metamizole (1st trim)
Live birth, 36 + 2, > 97th, f Polyhydramnios, perinatal death, tracheal atresia, normal karyotype (46, XX)
IvR2 29 years/18.4 Tachycardia, syncopes, pacemaker Ivabradine (1st trim)
Bisoprolol (0–35 + 5)
Levetiracetam (0–35 + 5, grand mal epilepsy)
Live birth, 35 + 5, < 10th, f IUGR (since GW 30), oligohydramnios (since GW 34), Caesarean section (vaginal bleeding)
IvR3 29 years/not available Tachycardia Ivabradine (0–41 + 3)
Bisoprolol (0–41 + 3)
Live birth, 41 + 3
> 10th, m
Uncomplicated pregnancy
BMI, body mass index; f, female; GW, gestational week; IUGR, intrauterine growth restriction; m, male; trim, trimester
The other two retrospective cases include a woman with comorbidities, intrauterine growth restriction, and premature birth (IvR2). The second was a healthy newborn (IvR3) after treatment with ivabradine throughout pregnancy. This mother reported a previous also uneventful pregnancy course under ivabradine.
Discussion
Pregnancy outcome
In our prospective cohort, there was only one neonate with major birth defects (cleft palate and atrial septal defect) from a diabetic mother with supraventricular tachycardia. No further major or cardiac malformations were reported among the 32 live births and 6 pregnancy losses. Cardiac defects have been observed in preclinical experimental studies. As a possible teratogenic mechanism, ivabradine-induced embryonic heart rate reduction leading to reduced cardiac output and hypoxia was discussed [16]. With a prevalence of almost 1 of 100 cardiac malformations belong to the most frequent birth defects [17, 18]. In our case, the neonate was large for gestational age and had a septum hypertrophy, supporting the assumption that maternal pre-gestational diabetes may also be causative for the observed cardiac defect and cleft palate. A causal relationship between ivabradine and cleft palate is unlikely, because ivabradine was discontinued at GW 7 + 5 before the beginning of the critical time period for the development of cleft palate at gestational week 9. The rate of major birth defects in our prospective cohort is comparable to the expected background rate of 2–3% [17, 18]. However, the limited number of pregnancies exposed during the entire period of embryogenesis does not allow excluding an increased risk, so far.
There was only one retrospectively reported case with a birth defect (malformation of the trachea) suspected as potentially ivabradine related. The concomitant exposures with naproxen and metamizole have not been evidenced as teratogens in human, whereas the high maternal BMI in the present case poses a risk for birth defects.
Taken together, we did not observe a specific pattern of congenital anomalies in our study. In addition, the presence of maternal comorbidities like diabetes and high BMI may have contributed to the observed birth defects.
The number of spontaneous abortions (3/38) was not higher than expected. The rate of preeclampsia and preterm birth in our cohort was within the expected range [19].
Treatment indications
Although the approved treatment indications for ivabradine, chronic heart failure and chronic stable angina pectoris (see Fig. 2 for changes of licenced use during the study period), are rare in women of reproductive age, we have observed an increasing need for information for off-label use in (younger) women considering pregnancy.
In our case series, most women were treated (off-label) for supraventricular tachycardia using lower doses than recommended (5–7.5 mg twice daily). Supraventricular tachycardia is not uncommon, especially in younger women, and is expected in 0.02–0.5% of pregnancies [20]. Aggravation of symptoms during pregnancy is described in about 20–50% of women related to various mechanisms, including pregnancy-related increase in cardiac heart rate and output and hormonal changes [20].
In 2015, the HRS (Heart Rhythm Society) considered ivabradine as promising for treating patients with inappropriate sinus tachycardia (IST) [13], followed by a recommendation for ivabradine by the ACC (American College of Cardiology) in 2016 [14]—though only one small randomized controlled study has yet supported beneficial effects [21]. Although still not approved for this indication, recent ESC guidelines [15] also recommend ivabradine alone or in combination with a beta-blocker in symptomatic patients with IST, for the treatment of focal atrial tachycardia or for postural orthostatic tachycardia syndrome (POTS), a condition considered as common cause for orthostatic intolerance in women between 15 and 25 years [20, 22] (Fig. 2).
Beta-blockers were required in 18% of women in our prospective cohort (Fig. 3). Discontinuation of ivabradine or switching to beta-blockers after recognition of pregnancy was well tolerated in most of our patients. Thus, necessity of ivabradine treatment should be critically considered in women with supraventricular tachycardia planning pregnancy. Data from previous studies suggest that discontinuation of ivabradine is possible in many (non-pregnant) patients with IST without recurrence. Even beneficial long-term effects after cessation of therapy are discussed [23, 24].
A high number of women in our cohort received therapy for anxiety and/or other psychiatric conditions. This is in line with findings from other studies reporting high rates of anxiety (25%) and depression (26%) in IST patients [25]. Anxiety is considered as non-cardiac symptom of inappropriate sinus tachycardia, and vice versa, anxiety may trigger tachycardia. Especially during pregnancy, empathic care and reassurance are necessary, to prevent increased concerns about possible harm to the fetus [26].
Preconception counselling is of utmost importance in all women considering pregnancy under medication [27]. Sinus tachycardia is a condition diagnosed mainly in young females and the first diagnosis is frequently made during or after pregnancy [25]. Considering the increasing off-label use of ivabradine as well as the high rate of > 40% unplanned pregnancies [28], more exposed pregnancies can be expected in the future. Well-established treatment options such as selected beta-blockers with low reproductive risk [29] should be considered in women of childbearing age before prescription of ivabradine.
Strength and limitations
A major strength of our study is the detailed exposure protocol and the multi-source ascertainment of pregnancy outcome via patients and their HCP followed by a case by case plausibility test.
Our data may not be representative of the German pregnant population; insofar, women with higher education are over-represented [30]. However, the regional distribution of enrolled patients is representative of the female population of childbearing age in Germany [11]. In addition, women on long-term ivabradine with severe cardiovascular conditions such as the approved treatment indications coronary artery disease and cardiac failure are under-represented in our cohort possibly favouring uneventful pregnancy outcomes. However, given the rarity of severe cardiac conditions in pregnant women, large study cohorts are not to be expected in the foreseeable future.
Conclusions
Our findings do not indicate that ivabradine is a major teratogen when used in early pregnancy. Although being the largest study published so far on the outcome of ivabradine-exposed pregnancies, the sample is too small to rule out embryotoxic effects. Further studies are needed to confirm or refute our findings.
In women planning pregnancy, established drugs of choice with low risk for the unborn should be preferred to ivabradine. For women with significant arrhythmias and severe cardiac disease, the impact of the underlying medical condition including comorbidities such as diabetes and high BMI on pregnancy outcome has to be considered when planning pregnancy.
After inadvertent ivabradine exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, all pregnant women with complicated arrhythmias or major cardiac disease should be carefully monitored.
Acknowledgements
We would like to thank all patients and physicians reporting pregnancies to the institute and acknowledge our colleagues from the German Embryotox Institute for counselling and thoroughly documenting cases, especially Sandra Zinke for continuous support with case documentation.
Contributions of authors
MH, ML, and CS designed the study. ML, EB, MH, and KD validated and analysed the data. ML, MH, and CS wrote the first draft of the manuscript, and all the authors critically revised subsequent manuscript drafts and contributed essential discussion points. All the authors approved the final version of this manuscript and they are responsible for the accuracy of this work.
Funding
Open Access funding enabled and organized by Projekt DEAL. This work was performed with financial support from the German Ministry of Health (BMG) and the German Federal Institute for Drugs and Medical Devices (BfArM). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
The study was registered with the German Clinical Trial register (DRKS00010502) and the study protocol was approved by the ethics committee of the Charité-Universitätsmedizin Berlin (EA1/107/16).
The study was registered with the German Clinical Trial register (DRKS00010502, 20.5.2016)
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | BISOPROLOL, IVABRADINE, METOPROLOL | DrugsGivenReaction | CC BY | 33501507 | 19,157,674 | 2021-07 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Foetal exposure during pregnancy'. | Ivabradine use in pregnant women-treatment indications and pregnancy outcome: an evaluation of the German Embryotox database.
OBJECTIVE
Ivabradine has been approved for the treatment of chronic heart failure and chronic stable angina pectoris in Europe. Based on adverse outcomes of reproductive animal studies and the lack of human data, ivabradine is considered contraindicated during pregnancy. The aim of this observational study is to analyse ivabradine use before and during pregnancy.
METHODS
We evaluated all ivabradine-related requests to the German Embryotox Institute from 2007 to 2019. Exposed pregnancies were analysed as to their outcome.
RESULTS
Off-label use for supraventricular tachycardia was frequent in women of childbearing age. Of 38 prospectively ascertained pregnancies with ivabradine exposure and completed follow-up, 32 resulted in live births, 3 in spontaneous abortions, and 3 were electively terminated. One neonate presented with major birth defects (atrial septal defect and cleft palate). In 33/38 patients, ivabradine was discontinued after confirmation of pregnancy without cardiac deterioration and 5/38 women continued ivabradine throughout pregnancy. In addition, there were 3 retrospectively reported pregnancies including one major birth defect (tracheal atresia).
CONCLUSIONS
This case series represents the largest cohort of ivabradine-exposed pregnancies, published so far. According to our findings, ivabradine appears not to be a major teratogen. However, established drugs of choice with strong evidence of low risk for the unborn should be preferred in women planning pregnancy. After inadvertent exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, close monitoring is necessary in pregnant women with supraventricular arrhythmias or cardiac disease.
Introduction
Ivabradine is a heart rate-lowering agent, acting through selective inhibition of the If (I-funny) channel without additional hemodynamic effects on cardiac function. The approved treatment indications in Europe are chronic heart failure and symptomatic treatment of chronic stable angina pectoris after failing or intolerance of beta-blockers [1].
Preclinical studies have suggested fetotoxicity and teratogenic effects. In pregnant rats, ivabradine exposure equivalent to therapeutic levels in humans was associated with increased intrauterine and postnatal mortality. Ventricular septal defects and complex anomalies of the great arteries were noted at doses 3 times the therapeutic human exposure. In pregnant rabbits, reduced fetal and placental weight as well as ectrodactyly was observed at doses 15 times the therapeutic human exposure [2]. In addition, dose-dependent mortality was seen in chicken and mice embryos [3, 4].
ESC guidelines consider ivabradine contraindicated during pregnancy and recommend discontinuation independent of treatment indication before conception with close clinical and echocardiographic monitoring [5].
Published data on ivabradine-exposed pregnancies are limited to three case reports, one with ivabradine treatment in the late first and early second trimester for sinus tachycardia after myocardial infarction [6], and the second starting ivabradine in the second trimester for deteriorating tachycardia-induced cardiomyopathy [7]. In the third case with first trimester exposure, pregnancy was electively terminated [8].
The limited human experience along with the teratogenic effects in animal studies makes it difficult to counsel inadvertently exposed pregnant women and their health care providers (HCP). Therefore, we decided to evaluate ivabradine-exposed pregnancies recorded by the German Embryotox Institute.
Methods
Study design
The German Embryotox Institute offers risk assessment on drug exposure during pregnancy to HCPs and patients. Up to 15,000 annual requests are answered, and in approximately 3500 critically exposed cases per year, the course and outcome of pregnancy are documented. At the initial contact, maternal characteristics and detailed exposure assessment are asked for after informed consent. About 8 weeks after the estimated date of birth, information on course and outcome of pregnancy is collected via structured telephone interview and mailed questionnaires. After a case by case plausibility check, additional information and health care records are requested in cases of adverse outcomes or inconsistent information.
Weeks of pregnancy were based on ultrasound and if not available on the last menstrual period (LMP). First trimester was defined as gestational week 2 + 0 (conception) until 12 + 6. Major birth defects were classified according to EUROCAT [9]. In prospectively ascertained pregnancies, neither the outcome of pregnancy nor prenatal pathology was known at the time of first contact. Pregnancies that were reported after birth or because of prenatal pathology were considered retrospective and evaluated separately. A detailed description of methodology adapted to the recommendations of the Strengthening of the Reporting of Observational studies in Epidemiology (STROBE) statement is given in Schaefer et al. [10] and Dathe and Schaefer [11].
We evaluated all requests on ivabradine to the German Embryotox Institute from 2007 to 2019.
Statistical analyses
Descriptive statistics were applied. For age, BMI, gestational weeks, neonatal weight, length, and head circumference, median and interquartile ranges are presented. Birth weights were adjusted to sex and gestational age at birth and percentile categories and standard deviation scores (SDS) were calculated according to the German perinatal survey [12].
Results
During the study period 2007–2019, our institute received 97 requests for ivabradine, 56 (58%) were from HCP, and 41 from patients. For further details, see Fig. 1. Treatment indications for ivabradine over time in relation to formal approval status and pregnancy labelling are summarized in Fig. 2. Supraventricular tachycardia was the most common reason for ivabradine treatment in our cohort. The first ivabradine-exposed pregnancy was recorded in March 2007, 1 year after approval and marketing authorization in Germany in January 2006.Fig. 1 Overview on information requests to the German Embryotox institute on ivabradine
Fig. 2 Treatment indication for ivabradine in 97 Embryotox requests over time in relation to pregnancy labelling and approval status by EMA/FDA. Recommendations for use of ivabradine in supraventricular tachycardias, i.e. inappropriate sinus tachycardia and postural orthostatic tachycardia syndrome: HRS/ESC 2015 [13], ACC/AHA/HFSA 2016 [14], ESC 2019 [15]. ACC, American College of Cardiology; EMA, European Medicines Agency; ESC, European Society of Cardiology; FDA, US Food and Drug Administration; HRS, Heart Rhythm Society
Fifty-six of the 97 requests were related to maternal exposure during pregnancy (Fig. 1). In 41 of these cases, follow-up on pregnancy outcome could be completed: 38 pregnancies were prospectively ascertained and three were retrospective reports.
Evaluation of prospectively ascertained pregnancies
Maternal characteristics
The median age of pregnant women was 29.5 years and the median BMI 22.5. Maternal characteristics are presented in detail in Table 1. The mean time at first contact was gestational week (GW) 7. Only 16 of the 38 prospectively ascertained pregnancies were actually planned (42%) and only 7 women (18%) started folic acid before conception. Contraceptive methods, mostly oral contraceptives (n = 4), were still used by 6 women at conception.Table 1 Maternal characteristics and pregnancy outcomes of ivabradine exposed pregnancies. Data are n (%) except for age, BMI, gestational week at birth, neonatal weight, length and head circumference, which are median and interquartile range
Maternal characteristics N (%) or median (IQR)
Maternal characteristics Age (n=38) 29.5 (26.5-34)
BMI (n=34) 22.5 (19.4-25.6)
Smoking (n=37)
>5 cigarettes/day 7 (19%)
<=5 cigarettes/day 4 (11%)
Alcohol (n=36)
<=1 drink/day 2 (6%)
>1 drink/day 1 (3%)
GW at first contact (n=38) 7 (5.6-11.1)
Previous pregnancies (n=38)
0 17(45%)
1 8 (21%)
2 13 (34%)
Pregnancy outcome and neonatal characteristics N (%) or median (IQR)
Pregnancy outcome (n=38) Spontaneous abortion 3
ETOP 3
Live birth 32
Pregnancy complications (n=32, only live births) Gestational diabetes 5 (16%)
Pre-eclampsia 2 (6%)
Cesarean section 15 (46%)
Preterm birth (<GW 37) 3 (9%)
Neonatal characteristics (n=33, including one pair of twins) GW at birth (n=32) 39 (37.9-39.9)
Weight, g (n=32) 3300 (2935-3610)
Length, cm (n=31) 50 (49-52.5)
HC, cm (n=30) 35 (34-36)
SGA (n=32) 2 (6%)
BMI, Body Mass Index; ETOP, elective termination of pregnancy; GW, gestational week; HC, head circumference; IQR, interquartile range; SGA, small for gestational age (<10th percentile)
Ivabradine exposure and co-medication
The most prevalent indication for ivabradine treatment was supraventricular tachycardia (n = 29) followed by heart failure (n = 3) and coronary artery disease (n = 2). Other reported indications (n = 4) included QT prolongation and sick-sinus-syndrome, although both are regarded as contraindications for ivabradine. The median daily ivabradine dose was 7.5 mg (IQR 5–10, min–max 2.5–15, n = 33).
Nearly all women were treated with ivabradine during the first trimester (n = 37) and 32 stopped treatment after recognition of pregnancy (Fig. 3), at median gestational week 6.1 (interquartile range 5.4–7.8). Cardiac decompensation or complications were not described after cessation of treatment. Only 5 women continued ivabradine throughout pregnancy or until shortly before delivery. One woman started treatment after first trimester (GW 18) of her twin pregnancy due to symptomatic tachycardia under beta-blocker therapy (Iv34 in Fig. 3).Fig. 3 Ivabradine exposure and pregnancy outcome of prospectively (n = 38) and retrospectively (n = 3) ascertained pregnancies
Cardiovascular co-medication consisted of beta-blockers (mainly metoprolol) in 15 women, mostly concomitantly (n = 7) or in replacement of ivabradine (n = 6). Two women initiated ivabradine after discontinuation of beta-blockers (Iv30 and Iv34 in Fig. 3). Six women received other cardiovascular medication, three methyldopa or amlodipine for hypertension, two digitoxin or verapamil for tachyarrhythmia, and one sildenafil and eplerenone for cardiomyopathy with cardiac failure (Iv17).
Psychiatric co-medication was reported in 7/38 women (5 of these with an anxiety disorder) including the teratogenic antiepileptic topiramate (n = 1) and pregabalin (n = 3).
Pregnancy outcome
Of the 38 prospectively ascertained ivabradine-exposed pregnancies, 32 were live births (33 infants including one pair of twins) and three resulted in early spontaneous abortions (Iv10, Iv17, and Iv19 in Fig. 3). One of these occurred in a woman with cardiomyopathy (Iv17) and various co-medications including hydroxycarbamide for thrombocythemia until GW 7. Three pregnancies were electively terminated for social/psychological reasons. Further information on pregnancy complications and outcomes is given in Table 1 and Fig. 3.
Major birth defects were reported in one pregnancy (Iv24) with 15 mg ivabradine plus bisoprolol until GW 7 + 5, both replaced by metoprolol. Polyhydramnios and preeclampsia led to Caesarean section in GW 35 after premature rupture of membranes. The premature child was large for gestational age (> 97 percentile) and had a cleft palate, an atrial septal defect II (spontaneously closed at the age of 10 months), and a hypertrophic cardiac septum. The mother was treated for sinus tachycardia, in addition she had diabetes type I and a BMI of 30.
Neonatal characteristics
Characteristics of the 33 neonates are summarized in Table 1. Compared to the German perinatal survey, standardized birth weights were lower with a median SDS of − 0.36, corresponding to a reduction of birth weight of 149 g for singletons born in gestational week 40.
Two neonates were small for gestational age, one of the twins (exposure only during 2nd and 3rd trimester, Iv34) and one after maternal ivabradine exposure until GW 12 (Iv29). Two neonates were large for gestational age, both mothers were diabetic (Iv24, type I diabetes and Iv27, gestational diabetes).
Retrospective cases
During the study period three pregnancies were retrospectively reported (further details are given in Table 2 and Fig. 3). One premature neonate (IvR1) with tracheal atresia died perinatally. The mother had been treated with ivabradine, naproxen, and metamizole during the first trimester until recognition of pregnancy (no further details available).Table 2 Retrospectively reported pregnancies
Case Maternal age/BMI Treatment indication Ivabradine exposure, other co-medication (GW or trimester) Pregnancy outcome, GW, birth weight percentile, sex Pregnancy complications, birth defects, and additional findings
IvR1 31 years/31.2 Tachycardia Ivabradine (1st trim)
Naproxen (1st trim)
Metamizole (1st trim)
Live birth, 36 + 2, > 97th, f Polyhydramnios, perinatal death, tracheal atresia, normal karyotype (46, XX)
IvR2 29 years/18.4 Tachycardia, syncopes, pacemaker Ivabradine (1st trim)
Bisoprolol (0–35 + 5)
Levetiracetam (0–35 + 5, grand mal epilepsy)
Live birth, 35 + 5, < 10th, f IUGR (since GW 30), oligohydramnios (since GW 34), Caesarean section (vaginal bleeding)
IvR3 29 years/not available Tachycardia Ivabradine (0–41 + 3)
Bisoprolol (0–41 + 3)
Live birth, 41 + 3
> 10th, m
Uncomplicated pregnancy
BMI, body mass index; f, female; GW, gestational week; IUGR, intrauterine growth restriction; m, male; trim, trimester
The other two retrospective cases include a woman with comorbidities, intrauterine growth restriction, and premature birth (IvR2). The second was a healthy newborn (IvR3) after treatment with ivabradine throughout pregnancy. This mother reported a previous also uneventful pregnancy course under ivabradine.
Discussion
Pregnancy outcome
In our prospective cohort, there was only one neonate with major birth defects (cleft palate and atrial septal defect) from a diabetic mother with supraventricular tachycardia. No further major or cardiac malformations were reported among the 32 live births and 6 pregnancy losses. Cardiac defects have been observed in preclinical experimental studies. As a possible teratogenic mechanism, ivabradine-induced embryonic heart rate reduction leading to reduced cardiac output and hypoxia was discussed [16]. With a prevalence of almost 1 of 100 cardiac malformations belong to the most frequent birth defects [17, 18]. In our case, the neonate was large for gestational age and had a septum hypertrophy, supporting the assumption that maternal pre-gestational diabetes may also be causative for the observed cardiac defect and cleft palate. A causal relationship between ivabradine and cleft palate is unlikely, because ivabradine was discontinued at GW 7 + 5 before the beginning of the critical time period for the development of cleft palate at gestational week 9. The rate of major birth defects in our prospective cohort is comparable to the expected background rate of 2–3% [17, 18]. However, the limited number of pregnancies exposed during the entire period of embryogenesis does not allow excluding an increased risk, so far.
There was only one retrospectively reported case with a birth defect (malformation of the trachea) suspected as potentially ivabradine related. The concomitant exposures with naproxen and metamizole have not been evidenced as teratogens in human, whereas the high maternal BMI in the present case poses a risk for birth defects.
Taken together, we did not observe a specific pattern of congenital anomalies in our study. In addition, the presence of maternal comorbidities like diabetes and high BMI may have contributed to the observed birth defects.
The number of spontaneous abortions (3/38) was not higher than expected. The rate of preeclampsia and preterm birth in our cohort was within the expected range [19].
Treatment indications
Although the approved treatment indications for ivabradine, chronic heart failure and chronic stable angina pectoris (see Fig. 2 for changes of licenced use during the study period), are rare in women of reproductive age, we have observed an increasing need for information for off-label use in (younger) women considering pregnancy.
In our case series, most women were treated (off-label) for supraventricular tachycardia using lower doses than recommended (5–7.5 mg twice daily). Supraventricular tachycardia is not uncommon, especially in younger women, and is expected in 0.02–0.5% of pregnancies [20]. Aggravation of symptoms during pregnancy is described in about 20–50% of women related to various mechanisms, including pregnancy-related increase in cardiac heart rate and output and hormonal changes [20].
In 2015, the HRS (Heart Rhythm Society) considered ivabradine as promising for treating patients with inappropriate sinus tachycardia (IST) [13], followed by a recommendation for ivabradine by the ACC (American College of Cardiology) in 2016 [14]—though only one small randomized controlled study has yet supported beneficial effects [21]. Although still not approved for this indication, recent ESC guidelines [15] also recommend ivabradine alone or in combination with a beta-blocker in symptomatic patients with IST, for the treatment of focal atrial tachycardia or for postural orthostatic tachycardia syndrome (POTS), a condition considered as common cause for orthostatic intolerance in women between 15 and 25 years [20, 22] (Fig. 2).
Beta-blockers were required in 18% of women in our prospective cohort (Fig. 3). Discontinuation of ivabradine or switching to beta-blockers after recognition of pregnancy was well tolerated in most of our patients. Thus, necessity of ivabradine treatment should be critically considered in women with supraventricular tachycardia planning pregnancy. Data from previous studies suggest that discontinuation of ivabradine is possible in many (non-pregnant) patients with IST without recurrence. Even beneficial long-term effects after cessation of therapy are discussed [23, 24].
A high number of women in our cohort received therapy for anxiety and/or other psychiatric conditions. This is in line with findings from other studies reporting high rates of anxiety (25%) and depression (26%) in IST patients [25]. Anxiety is considered as non-cardiac symptom of inappropriate sinus tachycardia, and vice versa, anxiety may trigger tachycardia. Especially during pregnancy, empathic care and reassurance are necessary, to prevent increased concerns about possible harm to the fetus [26].
Preconception counselling is of utmost importance in all women considering pregnancy under medication [27]. Sinus tachycardia is a condition diagnosed mainly in young females and the first diagnosis is frequently made during or after pregnancy [25]. Considering the increasing off-label use of ivabradine as well as the high rate of > 40% unplanned pregnancies [28], more exposed pregnancies can be expected in the future. Well-established treatment options such as selected beta-blockers with low reproductive risk [29] should be considered in women of childbearing age before prescription of ivabradine.
Strength and limitations
A major strength of our study is the detailed exposure protocol and the multi-source ascertainment of pregnancy outcome via patients and their HCP followed by a case by case plausibility test.
Our data may not be representative of the German pregnant population; insofar, women with higher education are over-represented [30]. However, the regional distribution of enrolled patients is representative of the female population of childbearing age in Germany [11]. In addition, women on long-term ivabradine with severe cardiovascular conditions such as the approved treatment indications coronary artery disease and cardiac failure are under-represented in our cohort possibly favouring uneventful pregnancy outcomes. However, given the rarity of severe cardiac conditions in pregnant women, large study cohorts are not to be expected in the foreseeable future.
Conclusions
Our findings do not indicate that ivabradine is a major teratogen when used in early pregnancy. Although being the largest study published so far on the outcome of ivabradine-exposed pregnancies, the sample is too small to rule out embryotoxic effects. Further studies are needed to confirm or refute our findings.
In women planning pregnancy, established drugs of choice with low risk for the unborn should be preferred to ivabradine. For women with significant arrhythmias and severe cardiac disease, the impact of the underlying medical condition including comorbidities such as diabetes and high BMI on pregnancy outcome has to be considered when planning pregnancy.
After inadvertent ivabradine exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, all pregnant women with complicated arrhythmias or major cardiac disease should be carefully monitored.
Acknowledgements
We would like to thank all patients and physicians reporting pregnancies to the institute and acknowledge our colleagues from the German Embryotox Institute for counselling and thoroughly documenting cases, especially Sandra Zinke for continuous support with case documentation.
Contributions of authors
MH, ML, and CS designed the study. ML, EB, MH, and KD validated and analysed the data. ML, MH, and CS wrote the first draft of the manuscript, and all the authors critically revised subsequent manuscript drafts and contributed essential discussion points. All the authors approved the final version of this manuscript and they are responsible for the accuracy of this work.
Funding
Open Access funding enabled and organized by Projekt DEAL. This work was performed with financial support from the German Ministry of Health (BMG) and the German Federal Institute for Drugs and Medical Devices (BfArM). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
The study was registered with the German Clinical Trial register (DRKS00010502) and the study protocol was approved by the ethics committee of the Charité-Universitätsmedizin Berlin (EA1/107/16).
The study was registered with the German Clinical Trial register (DRKS00010502, 20.5.2016)
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | BISOPROLOL, IVABRADINE, METOPROLOL | DrugsGivenReaction | CC BY | 33501507 | 19,697,614 | 2021-07 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Large for dates baby'. | Ivabradine use in pregnant women-treatment indications and pregnancy outcome: an evaluation of the German Embryotox database.
OBJECTIVE
Ivabradine has been approved for the treatment of chronic heart failure and chronic stable angina pectoris in Europe. Based on adverse outcomes of reproductive animal studies and the lack of human data, ivabradine is considered contraindicated during pregnancy. The aim of this observational study is to analyse ivabradine use before and during pregnancy.
METHODS
We evaluated all ivabradine-related requests to the German Embryotox Institute from 2007 to 2019. Exposed pregnancies were analysed as to their outcome.
RESULTS
Off-label use for supraventricular tachycardia was frequent in women of childbearing age. Of 38 prospectively ascertained pregnancies with ivabradine exposure and completed follow-up, 32 resulted in live births, 3 in spontaneous abortions, and 3 were electively terminated. One neonate presented with major birth defects (atrial septal defect and cleft palate). In 33/38 patients, ivabradine was discontinued after confirmation of pregnancy without cardiac deterioration and 5/38 women continued ivabradine throughout pregnancy. In addition, there were 3 retrospectively reported pregnancies including one major birth defect (tracheal atresia).
CONCLUSIONS
This case series represents the largest cohort of ivabradine-exposed pregnancies, published so far. According to our findings, ivabradine appears not to be a major teratogen. However, established drugs of choice with strong evidence of low risk for the unborn should be preferred in women planning pregnancy. After inadvertent exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, close monitoring is necessary in pregnant women with supraventricular arrhythmias or cardiac disease.
Introduction
Ivabradine is a heart rate-lowering agent, acting through selective inhibition of the If (I-funny) channel without additional hemodynamic effects on cardiac function. The approved treatment indications in Europe are chronic heart failure and symptomatic treatment of chronic stable angina pectoris after failing or intolerance of beta-blockers [1].
Preclinical studies have suggested fetotoxicity and teratogenic effects. In pregnant rats, ivabradine exposure equivalent to therapeutic levels in humans was associated with increased intrauterine and postnatal mortality. Ventricular septal defects and complex anomalies of the great arteries were noted at doses 3 times the therapeutic human exposure. In pregnant rabbits, reduced fetal and placental weight as well as ectrodactyly was observed at doses 15 times the therapeutic human exposure [2]. In addition, dose-dependent mortality was seen in chicken and mice embryos [3, 4].
ESC guidelines consider ivabradine contraindicated during pregnancy and recommend discontinuation independent of treatment indication before conception with close clinical and echocardiographic monitoring [5].
Published data on ivabradine-exposed pregnancies are limited to three case reports, one with ivabradine treatment in the late first and early second trimester for sinus tachycardia after myocardial infarction [6], and the second starting ivabradine in the second trimester for deteriorating tachycardia-induced cardiomyopathy [7]. In the third case with first trimester exposure, pregnancy was electively terminated [8].
The limited human experience along with the teratogenic effects in animal studies makes it difficult to counsel inadvertently exposed pregnant women and their health care providers (HCP). Therefore, we decided to evaluate ivabradine-exposed pregnancies recorded by the German Embryotox Institute.
Methods
Study design
The German Embryotox Institute offers risk assessment on drug exposure during pregnancy to HCPs and patients. Up to 15,000 annual requests are answered, and in approximately 3500 critically exposed cases per year, the course and outcome of pregnancy are documented. At the initial contact, maternal characteristics and detailed exposure assessment are asked for after informed consent. About 8 weeks after the estimated date of birth, information on course and outcome of pregnancy is collected via structured telephone interview and mailed questionnaires. After a case by case plausibility check, additional information and health care records are requested in cases of adverse outcomes or inconsistent information.
Weeks of pregnancy were based on ultrasound and if not available on the last menstrual period (LMP). First trimester was defined as gestational week 2 + 0 (conception) until 12 + 6. Major birth defects were classified according to EUROCAT [9]. In prospectively ascertained pregnancies, neither the outcome of pregnancy nor prenatal pathology was known at the time of first contact. Pregnancies that were reported after birth or because of prenatal pathology were considered retrospective and evaluated separately. A detailed description of methodology adapted to the recommendations of the Strengthening of the Reporting of Observational studies in Epidemiology (STROBE) statement is given in Schaefer et al. [10] and Dathe and Schaefer [11].
We evaluated all requests on ivabradine to the German Embryotox Institute from 2007 to 2019.
Statistical analyses
Descriptive statistics were applied. For age, BMI, gestational weeks, neonatal weight, length, and head circumference, median and interquartile ranges are presented. Birth weights were adjusted to sex and gestational age at birth and percentile categories and standard deviation scores (SDS) were calculated according to the German perinatal survey [12].
Results
During the study period 2007–2019, our institute received 97 requests for ivabradine, 56 (58%) were from HCP, and 41 from patients. For further details, see Fig. 1. Treatment indications for ivabradine over time in relation to formal approval status and pregnancy labelling are summarized in Fig. 2. Supraventricular tachycardia was the most common reason for ivabradine treatment in our cohort. The first ivabradine-exposed pregnancy was recorded in March 2007, 1 year after approval and marketing authorization in Germany in January 2006.Fig. 1 Overview on information requests to the German Embryotox institute on ivabradine
Fig. 2 Treatment indication for ivabradine in 97 Embryotox requests over time in relation to pregnancy labelling and approval status by EMA/FDA. Recommendations for use of ivabradine in supraventricular tachycardias, i.e. inappropriate sinus tachycardia and postural orthostatic tachycardia syndrome: HRS/ESC 2015 [13], ACC/AHA/HFSA 2016 [14], ESC 2019 [15]. ACC, American College of Cardiology; EMA, European Medicines Agency; ESC, European Society of Cardiology; FDA, US Food and Drug Administration; HRS, Heart Rhythm Society
Fifty-six of the 97 requests were related to maternal exposure during pregnancy (Fig. 1). In 41 of these cases, follow-up on pregnancy outcome could be completed: 38 pregnancies were prospectively ascertained and three were retrospective reports.
Evaluation of prospectively ascertained pregnancies
Maternal characteristics
The median age of pregnant women was 29.5 years and the median BMI 22.5. Maternal characteristics are presented in detail in Table 1. The mean time at first contact was gestational week (GW) 7. Only 16 of the 38 prospectively ascertained pregnancies were actually planned (42%) and only 7 women (18%) started folic acid before conception. Contraceptive methods, mostly oral contraceptives (n = 4), were still used by 6 women at conception.Table 1 Maternal characteristics and pregnancy outcomes of ivabradine exposed pregnancies. Data are n (%) except for age, BMI, gestational week at birth, neonatal weight, length and head circumference, which are median and interquartile range
Maternal characteristics N (%) or median (IQR)
Maternal characteristics Age (n=38) 29.5 (26.5-34)
BMI (n=34) 22.5 (19.4-25.6)
Smoking (n=37)
>5 cigarettes/day 7 (19%)
<=5 cigarettes/day 4 (11%)
Alcohol (n=36)
<=1 drink/day 2 (6%)
>1 drink/day 1 (3%)
GW at first contact (n=38) 7 (5.6-11.1)
Previous pregnancies (n=38)
0 17(45%)
1 8 (21%)
2 13 (34%)
Pregnancy outcome and neonatal characteristics N (%) or median (IQR)
Pregnancy outcome (n=38) Spontaneous abortion 3
ETOP 3
Live birth 32
Pregnancy complications (n=32, only live births) Gestational diabetes 5 (16%)
Pre-eclampsia 2 (6%)
Cesarean section 15 (46%)
Preterm birth (<GW 37) 3 (9%)
Neonatal characteristics (n=33, including one pair of twins) GW at birth (n=32) 39 (37.9-39.9)
Weight, g (n=32) 3300 (2935-3610)
Length, cm (n=31) 50 (49-52.5)
HC, cm (n=30) 35 (34-36)
SGA (n=32) 2 (6%)
BMI, Body Mass Index; ETOP, elective termination of pregnancy; GW, gestational week; HC, head circumference; IQR, interquartile range; SGA, small for gestational age (<10th percentile)
Ivabradine exposure and co-medication
The most prevalent indication for ivabradine treatment was supraventricular tachycardia (n = 29) followed by heart failure (n = 3) and coronary artery disease (n = 2). Other reported indications (n = 4) included QT prolongation and sick-sinus-syndrome, although both are regarded as contraindications for ivabradine. The median daily ivabradine dose was 7.5 mg (IQR 5–10, min–max 2.5–15, n = 33).
Nearly all women were treated with ivabradine during the first trimester (n = 37) and 32 stopped treatment after recognition of pregnancy (Fig. 3), at median gestational week 6.1 (interquartile range 5.4–7.8). Cardiac decompensation or complications were not described after cessation of treatment. Only 5 women continued ivabradine throughout pregnancy or until shortly before delivery. One woman started treatment after first trimester (GW 18) of her twin pregnancy due to symptomatic tachycardia under beta-blocker therapy (Iv34 in Fig. 3).Fig. 3 Ivabradine exposure and pregnancy outcome of prospectively (n = 38) and retrospectively (n = 3) ascertained pregnancies
Cardiovascular co-medication consisted of beta-blockers (mainly metoprolol) in 15 women, mostly concomitantly (n = 7) or in replacement of ivabradine (n = 6). Two women initiated ivabradine after discontinuation of beta-blockers (Iv30 and Iv34 in Fig. 3). Six women received other cardiovascular medication, three methyldopa or amlodipine for hypertension, two digitoxin or verapamil for tachyarrhythmia, and one sildenafil and eplerenone for cardiomyopathy with cardiac failure (Iv17).
Psychiatric co-medication was reported in 7/38 women (5 of these with an anxiety disorder) including the teratogenic antiepileptic topiramate (n = 1) and pregabalin (n = 3).
Pregnancy outcome
Of the 38 prospectively ascertained ivabradine-exposed pregnancies, 32 were live births (33 infants including one pair of twins) and three resulted in early spontaneous abortions (Iv10, Iv17, and Iv19 in Fig. 3). One of these occurred in a woman with cardiomyopathy (Iv17) and various co-medications including hydroxycarbamide for thrombocythemia until GW 7. Three pregnancies were electively terminated for social/psychological reasons. Further information on pregnancy complications and outcomes is given in Table 1 and Fig. 3.
Major birth defects were reported in one pregnancy (Iv24) with 15 mg ivabradine plus bisoprolol until GW 7 + 5, both replaced by metoprolol. Polyhydramnios and preeclampsia led to Caesarean section in GW 35 after premature rupture of membranes. The premature child was large for gestational age (> 97 percentile) and had a cleft palate, an atrial septal defect II (spontaneously closed at the age of 10 months), and a hypertrophic cardiac septum. The mother was treated for sinus tachycardia, in addition she had diabetes type I and a BMI of 30.
Neonatal characteristics
Characteristics of the 33 neonates are summarized in Table 1. Compared to the German perinatal survey, standardized birth weights were lower with a median SDS of − 0.36, corresponding to a reduction of birth weight of 149 g for singletons born in gestational week 40.
Two neonates were small for gestational age, one of the twins (exposure only during 2nd and 3rd trimester, Iv34) and one after maternal ivabradine exposure until GW 12 (Iv29). Two neonates were large for gestational age, both mothers were diabetic (Iv24, type I diabetes and Iv27, gestational diabetes).
Retrospective cases
During the study period three pregnancies were retrospectively reported (further details are given in Table 2 and Fig. 3). One premature neonate (IvR1) with tracheal atresia died perinatally. The mother had been treated with ivabradine, naproxen, and metamizole during the first trimester until recognition of pregnancy (no further details available).Table 2 Retrospectively reported pregnancies
Case Maternal age/BMI Treatment indication Ivabradine exposure, other co-medication (GW or trimester) Pregnancy outcome, GW, birth weight percentile, sex Pregnancy complications, birth defects, and additional findings
IvR1 31 years/31.2 Tachycardia Ivabradine (1st trim)
Naproxen (1st trim)
Metamizole (1st trim)
Live birth, 36 + 2, > 97th, f Polyhydramnios, perinatal death, tracheal atresia, normal karyotype (46, XX)
IvR2 29 years/18.4 Tachycardia, syncopes, pacemaker Ivabradine (1st trim)
Bisoprolol (0–35 + 5)
Levetiracetam (0–35 + 5, grand mal epilepsy)
Live birth, 35 + 5, < 10th, f IUGR (since GW 30), oligohydramnios (since GW 34), Caesarean section (vaginal bleeding)
IvR3 29 years/not available Tachycardia Ivabradine (0–41 + 3)
Bisoprolol (0–41 + 3)
Live birth, 41 + 3
> 10th, m
Uncomplicated pregnancy
BMI, body mass index; f, female; GW, gestational week; IUGR, intrauterine growth restriction; m, male; trim, trimester
The other two retrospective cases include a woman with comorbidities, intrauterine growth restriction, and premature birth (IvR2). The second was a healthy newborn (IvR3) after treatment with ivabradine throughout pregnancy. This mother reported a previous also uneventful pregnancy course under ivabradine.
Discussion
Pregnancy outcome
In our prospective cohort, there was only one neonate with major birth defects (cleft palate and atrial septal defect) from a diabetic mother with supraventricular tachycardia. No further major or cardiac malformations were reported among the 32 live births and 6 pregnancy losses. Cardiac defects have been observed in preclinical experimental studies. As a possible teratogenic mechanism, ivabradine-induced embryonic heart rate reduction leading to reduced cardiac output and hypoxia was discussed [16]. With a prevalence of almost 1 of 100 cardiac malformations belong to the most frequent birth defects [17, 18]. In our case, the neonate was large for gestational age and had a septum hypertrophy, supporting the assumption that maternal pre-gestational diabetes may also be causative for the observed cardiac defect and cleft palate. A causal relationship between ivabradine and cleft palate is unlikely, because ivabradine was discontinued at GW 7 + 5 before the beginning of the critical time period for the development of cleft palate at gestational week 9. The rate of major birth defects in our prospective cohort is comparable to the expected background rate of 2–3% [17, 18]. However, the limited number of pregnancies exposed during the entire period of embryogenesis does not allow excluding an increased risk, so far.
There was only one retrospectively reported case with a birth defect (malformation of the trachea) suspected as potentially ivabradine related. The concomitant exposures with naproxen and metamizole have not been evidenced as teratogens in human, whereas the high maternal BMI in the present case poses a risk for birth defects.
Taken together, we did not observe a specific pattern of congenital anomalies in our study. In addition, the presence of maternal comorbidities like diabetes and high BMI may have contributed to the observed birth defects.
The number of spontaneous abortions (3/38) was not higher than expected. The rate of preeclampsia and preterm birth in our cohort was within the expected range [19].
Treatment indications
Although the approved treatment indications for ivabradine, chronic heart failure and chronic stable angina pectoris (see Fig. 2 for changes of licenced use during the study period), are rare in women of reproductive age, we have observed an increasing need for information for off-label use in (younger) women considering pregnancy.
In our case series, most women were treated (off-label) for supraventricular tachycardia using lower doses than recommended (5–7.5 mg twice daily). Supraventricular tachycardia is not uncommon, especially in younger women, and is expected in 0.02–0.5% of pregnancies [20]. Aggravation of symptoms during pregnancy is described in about 20–50% of women related to various mechanisms, including pregnancy-related increase in cardiac heart rate and output and hormonal changes [20].
In 2015, the HRS (Heart Rhythm Society) considered ivabradine as promising for treating patients with inappropriate sinus tachycardia (IST) [13], followed by a recommendation for ivabradine by the ACC (American College of Cardiology) in 2016 [14]—though only one small randomized controlled study has yet supported beneficial effects [21]. Although still not approved for this indication, recent ESC guidelines [15] also recommend ivabradine alone or in combination with a beta-blocker in symptomatic patients with IST, for the treatment of focal atrial tachycardia or for postural orthostatic tachycardia syndrome (POTS), a condition considered as common cause for orthostatic intolerance in women between 15 and 25 years [20, 22] (Fig. 2).
Beta-blockers were required in 18% of women in our prospective cohort (Fig. 3). Discontinuation of ivabradine or switching to beta-blockers after recognition of pregnancy was well tolerated in most of our patients. Thus, necessity of ivabradine treatment should be critically considered in women with supraventricular tachycardia planning pregnancy. Data from previous studies suggest that discontinuation of ivabradine is possible in many (non-pregnant) patients with IST without recurrence. Even beneficial long-term effects after cessation of therapy are discussed [23, 24].
A high number of women in our cohort received therapy for anxiety and/or other psychiatric conditions. This is in line with findings from other studies reporting high rates of anxiety (25%) and depression (26%) in IST patients [25]. Anxiety is considered as non-cardiac symptom of inappropriate sinus tachycardia, and vice versa, anxiety may trigger tachycardia. Especially during pregnancy, empathic care and reassurance are necessary, to prevent increased concerns about possible harm to the fetus [26].
Preconception counselling is of utmost importance in all women considering pregnancy under medication [27]. Sinus tachycardia is a condition diagnosed mainly in young females and the first diagnosis is frequently made during or after pregnancy [25]. Considering the increasing off-label use of ivabradine as well as the high rate of > 40% unplanned pregnancies [28], more exposed pregnancies can be expected in the future. Well-established treatment options such as selected beta-blockers with low reproductive risk [29] should be considered in women of childbearing age before prescription of ivabradine.
Strength and limitations
A major strength of our study is the detailed exposure protocol and the multi-source ascertainment of pregnancy outcome via patients and their HCP followed by a case by case plausibility test.
Our data may not be representative of the German pregnant population; insofar, women with higher education are over-represented [30]. However, the regional distribution of enrolled patients is representative of the female population of childbearing age in Germany [11]. In addition, women on long-term ivabradine with severe cardiovascular conditions such as the approved treatment indications coronary artery disease and cardiac failure are under-represented in our cohort possibly favouring uneventful pregnancy outcomes. However, given the rarity of severe cardiac conditions in pregnant women, large study cohorts are not to be expected in the foreseeable future.
Conclusions
Our findings do not indicate that ivabradine is a major teratogen when used in early pregnancy. Although being the largest study published so far on the outcome of ivabradine-exposed pregnancies, the sample is too small to rule out embryotoxic effects. Further studies are needed to confirm or refute our findings.
In women planning pregnancy, established drugs of choice with low risk for the unborn should be preferred to ivabradine. For women with significant arrhythmias and severe cardiac disease, the impact of the underlying medical condition including comorbidities such as diabetes and high BMI on pregnancy outcome has to be considered when planning pregnancy.
After inadvertent ivabradine exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, all pregnant women with complicated arrhythmias or major cardiac disease should be carefully monitored.
Acknowledgements
We would like to thank all patients and physicians reporting pregnancies to the institute and acknowledge our colleagues from the German Embryotox Institute for counselling and thoroughly documenting cases, especially Sandra Zinke for continuous support with case documentation.
Contributions of authors
MH, ML, and CS designed the study. ML, EB, MH, and KD validated and analysed the data. ML, MH, and CS wrote the first draft of the manuscript, and all the authors critically revised subsequent manuscript drafts and contributed essential discussion points. All the authors approved the final version of this manuscript and they are responsible for the accuracy of this work.
Funding
Open Access funding enabled and organized by Projekt DEAL. This work was performed with financial support from the German Ministry of Health (BMG) and the German Federal Institute for Drugs and Medical Devices (BfArM). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
The study was registered with the German Clinical Trial register (DRKS00010502) and the study protocol was approved by the ethics committee of the Charité-Universitätsmedizin Berlin (EA1/107/16).
The study was registered with the German Clinical Trial register (DRKS00010502, 20.5.2016)
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | BISOPROLOL, IVABRADINE, METOPROLOL | DrugsGivenReaction | CC BY | 33501507 | 19,697,614 | 2021-07 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Polyhydramnios'. | Ivabradine use in pregnant women-treatment indications and pregnancy outcome: an evaluation of the German Embryotox database.
OBJECTIVE
Ivabradine has been approved for the treatment of chronic heart failure and chronic stable angina pectoris in Europe. Based on adverse outcomes of reproductive animal studies and the lack of human data, ivabradine is considered contraindicated during pregnancy. The aim of this observational study is to analyse ivabradine use before and during pregnancy.
METHODS
We evaluated all ivabradine-related requests to the German Embryotox Institute from 2007 to 2019. Exposed pregnancies were analysed as to their outcome.
RESULTS
Off-label use for supraventricular tachycardia was frequent in women of childbearing age. Of 38 prospectively ascertained pregnancies with ivabradine exposure and completed follow-up, 32 resulted in live births, 3 in spontaneous abortions, and 3 were electively terminated. One neonate presented with major birth defects (atrial septal defect and cleft palate). In 33/38 patients, ivabradine was discontinued after confirmation of pregnancy without cardiac deterioration and 5/38 women continued ivabradine throughout pregnancy. In addition, there were 3 retrospectively reported pregnancies including one major birth defect (tracheal atresia).
CONCLUSIONS
This case series represents the largest cohort of ivabradine-exposed pregnancies, published so far. According to our findings, ivabradine appears not to be a major teratogen. However, established drugs of choice with strong evidence of low risk for the unborn should be preferred in women planning pregnancy. After inadvertent exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, close monitoring is necessary in pregnant women with supraventricular arrhythmias or cardiac disease.
Introduction
Ivabradine is a heart rate-lowering agent, acting through selective inhibition of the If (I-funny) channel without additional hemodynamic effects on cardiac function. The approved treatment indications in Europe are chronic heart failure and symptomatic treatment of chronic stable angina pectoris after failing or intolerance of beta-blockers [1].
Preclinical studies have suggested fetotoxicity and teratogenic effects. In pregnant rats, ivabradine exposure equivalent to therapeutic levels in humans was associated with increased intrauterine and postnatal mortality. Ventricular septal defects and complex anomalies of the great arteries were noted at doses 3 times the therapeutic human exposure. In pregnant rabbits, reduced fetal and placental weight as well as ectrodactyly was observed at doses 15 times the therapeutic human exposure [2]. In addition, dose-dependent mortality was seen in chicken and mice embryos [3, 4].
ESC guidelines consider ivabradine contraindicated during pregnancy and recommend discontinuation independent of treatment indication before conception with close clinical and echocardiographic monitoring [5].
Published data on ivabradine-exposed pregnancies are limited to three case reports, one with ivabradine treatment in the late first and early second trimester for sinus tachycardia after myocardial infarction [6], and the second starting ivabradine in the second trimester for deteriorating tachycardia-induced cardiomyopathy [7]. In the third case with first trimester exposure, pregnancy was electively terminated [8].
The limited human experience along with the teratogenic effects in animal studies makes it difficult to counsel inadvertently exposed pregnant women and their health care providers (HCP). Therefore, we decided to evaluate ivabradine-exposed pregnancies recorded by the German Embryotox Institute.
Methods
Study design
The German Embryotox Institute offers risk assessment on drug exposure during pregnancy to HCPs and patients. Up to 15,000 annual requests are answered, and in approximately 3500 critically exposed cases per year, the course and outcome of pregnancy are documented. At the initial contact, maternal characteristics and detailed exposure assessment are asked for after informed consent. About 8 weeks after the estimated date of birth, information on course and outcome of pregnancy is collected via structured telephone interview and mailed questionnaires. After a case by case plausibility check, additional information and health care records are requested in cases of adverse outcomes or inconsistent information.
Weeks of pregnancy were based on ultrasound and if not available on the last menstrual period (LMP). First trimester was defined as gestational week 2 + 0 (conception) until 12 + 6. Major birth defects were classified according to EUROCAT [9]. In prospectively ascertained pregnancies, neither the outcome of pregnancy nor prenatal pathology was known at the time of first contact. Pregnancies that were reported after birth or because of prenatal pathology were considered retrospective and evaluated separately. A detailed description of methodology adapted to the recommendations of the Strengthening of the Reporting of Observational studies in Epidemiology (STROBE) statement is given in Schaefer et al. [10] and Dathe and Schaefer [11].
We evaluated all requests on ivabradine to the German Embryotox Institute from 2007 to 2019.
Statistical analyses
Descriptive statistics were applied. For age, BMI, gestational weeks, neonatal weight, length, and head circumference, median and interquartile ranges are presented. Birth weights were adjusted to sex and gestational age at birth and percentile categories and standard deviation scores (SDS) were calculated according to the German perinatal survey [12].
Results
During the study period 2007–2019, our institute received 97 requests for ivabradine, 56 (58%) were from HCP, and 41 from patients. For further details, see Fig. 1. Treatment indications for ivabradine over time in relation to formal approval status and pregnancy labelling are summarized in Fig. 2. Supraventricular tachycardia was the most common reason for ivabradine treatment in our cohort. The first ivabradine-exposed pregnancy was recorded in March 2007, 1 year after approval and marketing authorization in Germany in January 2006.Fig. 1 Overview on information requests to the German Embryotox institute on ivabradine
Fig. 2 Treatment indication for ivabradine in 97 Embryotox requests over time in relation to pregnancy labelling and approval status by EMA/FDA. Recommendations for use of ivabradine in supraventricular tachycardias, i.e. inappropriate sinus tachycardia and postural orthostatic tachycardia syndrome: HRS/ESC 2015 [13], ACC/AHA/HFSA 2016 [14], ESC 2019 [15]. ACC, American College of Cardiology; EMA, European Medicines Agency; ESC, European Society of Cardiology; FDA, US Food and Drug Administration; HRS, Heart Rhythm Society
Fifty-six of the 97 requests were related to maternal exposure during pregnancy (Fig. 1). In 41 of these cases, follow-up on pregnancy outcome could be completed: 38 pregnancies were prospectively ascertained and three were retrospective reports.
Evaluation of prospectively ascertained pregnancies
Maternal characteristics
The median age of pregnant women was 29.5 years and the median BMI 22.5. Maternal characteristics are presented in detail in Table 1. The mean time at first contact was gestational week (GW) 7. Only 16 of the 38 prospectively ascertained pregnancies were actually planned (42%) and only 7 women (18%) started folic acid before conception. Contraceptive methods, mostly oral contraceptives (n = 4), were still used by 6 women at conception.Table 1 Maternal characteristics and pregnancy outcomes of ivabradine exposed pregnancies. Data are n (%) except for age, BMI, gestational week at birth, neonatal weight, length and head circumference, which are median and interquartile range
Maternal characteristics N (%) or median (IQR)
Maternal characteristics Age (n=38) 29.5 (26.5-34)
BMI (n=34) 22.5 (19.4-25.6)
Smoking (n=37)
>5 cigarettes/day 7 (19%)
<=5 cigarettes/day 4 (11%)
Alcohol (n=36)
<=1 drink/day 2 (6%)
>1 drink/day 1 (3%)
GW at first contact (n=38) 7 (5.6-11.1)
Previous pregnancies (n=38)
0 17(45%)
1 8 (21%)
2 13 (34%)
Pregnancy outcome and neonatal characteristics N (%) or median (IQR)
Pregnancy outcome (n=38) Spontaneous abortion 3
ETOP 3
Live birth 32
Pregnancy complications (n=32, only live births) Gestational diabetes 5 (16%)
Pre-eclampsia 2 (6%)
Cesarean section 15 (46%)
Preterm birth (<GW 37) 3 (9%)
Neonatal characteristics (n=33, including one pair of twins) GW at birth (n=32) 39 (37.9-39.9)
Weight, g (n=32) 3300 (2935-3610)
Length, cm (n=31) 50 (49-52.5)
HC, cm (n=30) 35 (34-36)
SGA (n=32) 2 (6%)
BMI, Body Mass Index; ETOP, elective termination of pregnancy; GW, gestational week; HC, head circumference; IQR, interquartile range; SGA, small for gestational age (<10th percentile)
Ivabradine exposure and co-medication
The most prevalent indication for ivabradine treatment was supraventricular tachycardia (n = 29) followed by heart failure (n = 3) and coronary artery disease (n = 2). Other reported indications (n = 4) included QT prolongation and sick-sinus-syndrome, although both are regarded as contraindications for ivabradine. The median daily ivabradine dose was 7.5 mg (IQR 5–10, min–max 2.5–15, n = 33).
Nearly all women were treated with ivabradine during the first trimester (n = 37) and 32 stopped treatment after recognition of pregnancy (Fig. 3), at median gestational week 6.1 (interquartile range 5.4–7.8). Cardiac decompensation or complications were not described after cessation of treatment. Only 5 women continued ivabradine throughout pregnancy or until shortly before delivery. One woman started treatment after first trimester (GW 18) of her twin pregnancy due to symptomatic tachycardia under beta-blocker therapy (Iv34 in Fig. 3).Fig. 3 Ivabradine exposure and pregnancy outcome of prospectively (n = 38) and retrospectively (n = 3) ascertained pregnancies
Cardiovascular co-medication consisted of beta-blockers (mainly metoprolol) in 15 women, mostly concomitantly (n = 7) or in replacement of ivabradine (n = 6). Two women initiated ivabradine after discontinuation of beta-blockers (Iv30 and Iv34 in Fig. 3). Six women received other cardiovascular medication, three methyldopa or amlodipine for hypertension, two digitoxin or verapamil for tachyarrhythmia, and one sildenafil and eplerenone for cardiomyopathy with cardiac failure (Iv17).
Psychiatric co-medication was reported in 7/38 women (5 of these with an anxiety disorder) including the teratogenic antiepileptic topiramate (n = 1) and pregabalin (n = 3).
Pregnancy outcome
Of the 38 prospectively ascertained ivabradine-exposed pregnancies, 32 were live births (33 infants including one pair of twins) and three resulted in early spontaneous abortions (Iv10, Iv17, and Iv19 in Fig. 3). One of these occurred in a woman with cardiomyopathy (Iv17) and various co-medications including hydroxycarbamide for thrombocythemia until GW 7. Three pregnancies were electively terminated for social/psychological reasons. Further information on pregnancy complications and outcomes is given in Table 1 and Fig. 3.
Major birth defects were reported in one pregnancy (Iv24) with 15 mg ivabradine plus bisoprolol until GW 7 + 5, both replaced by metoprolol. Polyhydramnios and preeclampsia led to Caesarean section in GW 35 after premature rupture of membranes. The premature child was large for gestational age (> 97 percentile) and had a cleft palate, an atrial septal defect II (spontaneously closed at the age of 10 months), and a hypertrophic cardiac septum. The mother was treated for sinus tachycardia, in addition she had diabetes type I and a BMI of 30.
Neonatal characteristics
Characteristics of the 33 neonates are summarized in Table 1. Compared to the German perinatal survey, standardized birth weights were lower with a median SDS of − 0.36, corresponding to a reduction of birth weight of 149 g for singletons born in gestational week 40.
Two neonates were small for gestational age, one of the twins (exposure only during 2nd and 3rd trimester, Iv34) and one after maternal ivabradine exposure until GW 12 (Iv29). Two neonates were large for gestational age, both mothers were diabetic (Iv24, type I diabetes and Iv27, gestational diabetes).
Retrospective cases
During the study period three pregnancies were retrospectively reported (further details are given in Table 2 and Fig. 3). One premature neonate (IvR1) with tracheal atresia died perinatally. The mother had been treated with ivabradine, naproxen, and metamizole during the first trimester until recognition of pregnancy (no further details available).Table 2 Retrospectively reported pregnancies
Case Maternal age/BMI Treatment indication Ivabradine exposure, other co-medication (GW or trimester) Pregnancy outcome, GW, birth weight percentile, sex Pregnancy complications, birth defects, and additional findings
IvR1 31 years/31.2 Tachycardia Ivabradine (1st trim)
Naproxen (1st trim)
Metamizole (1st trim)
Live birth, 36 + 2, > 97th, f Polyhydramnios, perinatal death, tracheal atresia, normal karyotype (46, XX)
IvR2 29 years/18.4 Tachycardia, syncopes, pacemaker Ivabradine (1st trim)
Bisoprolol (0–35 + 5)
Levetiracetam (0–35 + 5, grand mal epilepsy)
Live birth, 35 + 5, < 10th, f IUGR (since GW 30), oligohydramnios (since GW 34), Caesarean section (vaginal bleeding)
IvR3 29 years/not available Tachycardia Ivabradine (0–41 + 3)
Bisoprolol (0–41 + 3)
Live birth, 41 + 3
> 10th, m
Uncomplicated pregnancy
BMI, body mass index; f, female; GW, gestational week; IUGR, intrauterine growth restriction; m, male; trim, trimester
The other two retrospective cases include a woman with comorbidities, intrauterine growth restriction, and premature birth (IvR2). The second was a healthy newborn (IvR3) after treatment with ivabradine throughout pregnancy. This mother reported a previous also uneventful pregnancy course under ivabradine.
Discussion
Pregnancy outcome
In our prospective cohort, there was only one neonate with major birth defects (cleft palate and atrial septal defect) from a diabetic mother with supraventricular tachycardia. No further major or cardiac malformations were reported among the 32 live births and 6 pregnancy losses. Cardiac defects have been observed in preclinical experimental studies. As a possible teratogenic mechanism, ivabradine-induced embryonic heart rate reduction leading to reduced cardiac output and hypoxia was discussed [16]. With a prevalence of almost 1 of 100 cardiac malformations belong to the most frequent birth defects [17, 18]. In our case, the neonate was large for gestational age and had a septum hypertrophy, supporting the assumption that maternal pre-gestational diabetes may also be causative for the observed cardiac defect and cleft palate. A causal relationship between ivabradine and cleft palate is unlikely, because ivabradine was discontinued at GW 7 + 5 before the beginning of the critical time period for the development of cleft palate at gestational week 9. The rate of major birth defects in our prospective cohort is comparable to the expected background rate of 2–3% [17, 18]. However, the limited number of pregnancies exposed during the entire period of embryogenesis does not allow excluding an increased risk, so far.
There was only one retrospectively reported case with a birth defect (malformation of the trachea) suspected as potentially ivabradine related. The concomitant exposures with naproxen and metamizole have not been evidenced as teratogens in human, whereas the high maternal BMI in the present case poses a risk for birth defects.
Taken together, we did not observe a specific pattern of congenital anomalies in our study. In addition, the presence of maternal comorbidities like diabetes and high BMI may have contributed to the observed birth defects.
The number of spontaneous abortions (3/38) was not higher than expected. The rate of preeclampsia and preterm birth in our cohort was within the expected range [19].
Treatment indications
Although the approved treatment indications for ivabradine, chronic heart failure and chronic stable angina pectoris (see Fig. 2 for changes of licenced use during the study period), are rare in women of reproductive age, we have observed an increasing need for information for off-label use in (younger) women considering pregnancy.
In our case series, most women were treated (off-label) for supraventricular tachycardia using lower doses than recommended (5–7.5 mg twice daily). Supraventricular tachycardia is not uncommon, especially in younger women, and is expected in 0.02–0.5% of pregnancies [20]. Aggravation of symptoms during pregnancy is described in about 20–50% of women related to various mechanisms, including pregnancy-related increase in cardiac heart rate and output and hormonal changes [20].
In 2015, the HRS (Heart Rhythm Society) considered ivabradine as promising for treating patients with inappropriate sinus tachycardia (IST) [13], followed by a recommendation for ivabradine by the ACC (American College of Cardiology) in 2016 [14]—though only one small randomized controlled study has yet supported beneficial effects [21]. Although still not approved for this indication, recent ESC guidelines [15] also recommend ivabradine alone or in combination with a beta-blocker in symptomatic patients with IST, for the treatment of focal atrial tachycardia or for postural orthostatic tachycardia syndrome (POTS), a condition considered as common cause for orthostatic intolerance in women between 15 and 25 years [20, 22] (Fig. 2).
Beta-blockers were required in 18% of women in our prospective cohort (Fig. 3). Discontinuation of ivabradine or switching to beta-blockers after recognition of pregnancy was well tolerated in most of our patients. Thus, necessity of ivabradine treatment should be critically considered in women with supraventricular tachycardia planning pregnancy. Data from previous studies suggest that discontinuation of ivabradine is possible in many (non-pregnant) patients with IST without recurrence. Even beneficial long-term effects after cessation of therapy are discussed [23, 24].
A high number of women in our cohort received therapy for anxiety and/or other psychiatric conditions. This is in line with findings from other studies reporting high rates of anxiety (25%) and depression (26%) in IST patients [25]. Anxiety is considered as non-cardiac symptom of inappropriate sinus tachycardia, and vice versa, anxiety may trigger tachycardia. Especially during pregnancy, empathic care and reassurance are necessary, to prevent increased concerns about possible harm to the fetus [26].
Preconception counselling is of utmost importance in all women considering pregnancy under medication [27]. Sinus tachycardia is a condition diagnosed mainly in young females and the first diagnosis is frequently made during or after pregnancy [25]. Considering the increasing off-label use of ivabradine as well as the high rate of > 40% unplanned pregnancies [28], more exposed pregnancies can be expected in the future. Well-established treatment options such as selected beta-blockers with low reproductive risk [29] should be considered in women of childbearing age before prescription of ivabradine.
Strength and limitations
A major strength of our study is the detailed exposure protocol and the multi-source ascertainment of pregnancy outcome via patients and their HCP followed by a case by case plausibility test.
Our data may not be representative of the German pregnant population; insofar, women with higher education are over-represented [30]. However, the regional distribution of enrolled patients is representative of the female population of childbearing age in Germany [11]. In addition, women on long-term ivabradine with severe cardiovascular conditions such as the approved treatment indications coronary artery disease and cardiac failure are under-represented in our cohort possibly favouring uneventful pregnancy outcomes. However, given the rarity of severe cardiac conditions in pregnant women, large study cohorts are not to be expected in the foreseeable future.
Conclusions
Our findings do not indicate that ivabradine is a major teratogen when used in early pregnancy. Although being the largest study published so far on the outcome of ivabradine-exposed pregnancies, the sample is too small to rule out embryotoxic effects. Further studies are needed to confirm or refute our findings.
In women planning pregnancy, established drugs of choice with low risk for the unborn should be preferred to ivabradine. For women with significant arrhythmias and severe cardiac disease, the impact of the underlying medical condition including comorbidities such as diabetes and high BMI on pregnancy outcome has to be considered when planning pregnancy.
After inadvertent ivabradine exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, all pregnant women with complicated arrhythmias or major cardiac disease should be carefully monitored.
Acknowledgements
We would like to thank all patients and physicians reporting pregnancies to the institute and acknowledge our colleagues from the German Embryotox Institute for counselling and thoroughly documenting cases, especially Sandra Zinke for continuous support with case documentation.
Contributions of authors
MH, ML, and CS designed the study. ML, EB, MH, and KD validated and analysed the data. ML, MH, and CS wrote the first draft of the manuscript, and all the authors critically revised subsequent manuscript drafts and contributed essential discussion points. All the authors approved the final version of this manuscript and they are responsible for the accuracy of this work.
Funding
Open Access funding enabled and organized by Projekt DEAL. This work was performed with financial support from the German Ministry of Health (BMG) and the German Federal Institute for Drugs and Medical Devices (BfArM). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
The study was registered with the German Clinical Trial register (DRKS00010502) and the study protocol was approved by the ethics committee of the Charité-Universitätsmedizin Berlin (EA1/107/16).
The study was registered with the German Clinical Trial register (DRKS00010502, 20.5.2016)
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | BISOPROLOL, IVABRADINE, METOPROLOL | DrugsGivenReaction | CC BY | 33501507 | 19,157,674 | 2021-07 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Pre-eclampsia'. | Ivabradine use in pregnant women-treatment indications and pregnancy outcome: an evaluation of the German Embryotox database.
OBJECTIVE
Ivabradine has been approved for the treatment of chronic heart failure and chronic stable angina pectoris in Europe. Based on adverse outcomes of reproductive animal studies and the lack of human data, ivabradine is considered contraindicated during pregnancy. The aim of this observational study is to analyse ivabradine use before and during pregnancy.
METHODS
We evaluated all ivabradine-related requests to the German Embryotox Institute from 2007 to 2019. Exposed pregnancies were analysed as to their outcome.
RESULTS
Off-label use for supraventricular tachycardia was frequent in women of childbearing age. Of 38 prospectively ascertained pregnancies with ivabradine exposure and completed follow-up, 32 resulted in live births, 3 in spontaneous abortions, and 3 were electively terminated. One neonate presented with major birth defects (atrial septal defect and cleft palate). In 33/38 patients, ivabradine was discontinued after confirmation of pregnancy without cardiac deterioration and 5/38 women continued ivabradine throughout pregnancy. In addition, there were 3 retrospectively reported pregnancies including one major birth defect (tracheal atresia).
CONCLUSIONS
This case series represents the largest cohort of ivabradine-exposed pregnancies, published so far. According to our findings, ivabradine appears not to be a major teratogen. However, established drugs of choice with strong evidence of low risk for the unborn should be preferred in women planning pregnancy. After inadvertent exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, close monitoring is necessary in pregnant women with supraventricular arrhythmias or cardiac disease.
Introduction
Ivabradine is a heart rate-lowering agent, acting through selective inhibition of the If (I-funny) channel without additional hemodynamic effects on cardiac function. The approved treatment indications in Europe are chronic heart failure and symptomatic treatment of chronic stable angina pectoris after failing or intolerance of beta-blockers [1].
Preclinical studies have suggested fetotoxicity and teratogenic effects. In pregnant rats, ivabradine exposure equivalent to therapeutic levels in humans was associated with increased intrauterine and postnatal mortality. Ventricular septal defects and complex anomalies of the great arteries were noted at doses 3 times the therapeutic human exposure. In pregnant rabbits, reduced fetal and placental weight as well as ectrodactyly was observed at doses 15 times the therapeutic human exposure [2]. In addition, dose-dependent mortality was seen in chicken and mice embryos [3, 4].
ESC guidelines consider ivabradine contraindicated during pregnancy and recommend discontinuation independent of treatment indication before conception with close clinical and echocardiographic monitoring [5].
Published data on ivabradine-exposed pregnancies are limited to three case reports, one with ivabradine treatment in the late first and early second trimester for sinus tachycardia after myocardial infarction [6], and the second starting ivabradine in the second trimester for deteriorating tachycardia-induced cardiomyopathy [7]. In the third case with first trimester exposure, pregnancy was electively terminated [8].
The limited human experience along with the teratogenic effects in animal studies makes it difficult to counsel inadvertently exposed pregnant women and their health care providers (HCP). Therefore, we decided to evaluate ivabradine-exposed pregnancies recorded by the German Embryotox Institute.
Methods
Study design
The German Embryotox Institute offers risk assessment on drug exposure during pregnancy to HCPs and patients. Up to 15,000 annual requests are answered, and in approximately 3500 critically exposed cases per year, the course and outcome of pregnancy are documented. At the initial contact, maternal characteristics and detailed exposure assessment are asked for after informed consent. About 8 weeks after the estimated date of birth, information on course and outcome of pregnancy is collected via structured telephone interview and mailed questionnaires. After a case by case plausibility check, additional information and health care records are requested in cases of adverse outcomes or inconsistent information.
Weeks of pregnancy were based on ultrasound and if not available on the last menstrual period (LMP). First trimester was defined as gestational week 2 + 0 (conception) until 12 + 6. Major birth defects were classified according to EUROCAT [9]. In prospectively ascertained pregnancies, neither the outcome of pregnancy nor prenatal pathology was known at the time of first contact. Pregnancies that were reported after birth or because of prenatal pathology were considered retrospective and evaluated separately. A detailed description of methodology adapted to the recommendations of the Strengthening of the Reporting of Observational studies in Epidemiology (STROBE) statement is given in Schaefer et al. [10] and Dathe and Schaefer [11].
We evaluated all requests on ivabradine to the German Embryotox Institute from 2007 to 2019.
Statistical analyses
Descriptive statistics were applied. For age, BMI, gestational weeks, neonatal weight, length, and head circumference, median and interquartile ranges are presented. Birth weights were adjusted to sex and gestational age at birth and percentile categories and standard deviation scores (SDS) were calculated according to the German perinatal survey [12].
Results
During the study period 2007–2019, our institute received 97 requests for ivabradine, 56 (58%) were from HCP, and 41 from patients. For further details, see Fig. 1. Treatment indications for ivabradine over time in relation to formal approval status and pregnancy labelling are summarized in Fig. 2. Supraventricular tachycardia was the most common reason for ivabradine treatment in our cohort. The first ivabradine-exposed pregnancy was recorded in March 2007, 1 year after approval and marketing authorization in Germany in January 2006.Fig. 1 Overview on information requests to the German Embryotox institute on ivabradine
Fig. 2 Treatment indication for ivabradine in 97 Embryotox requests over time in relation to pregnancy labelling and approval status by EMA/FDA. Recommendations for use of ivabradine in supraventricular tachycardias, i.e. inappropriate sinus tachycardia and postural orthostatic tachycardia syndrome: HRS/ESC 2015 [13], ACC/AHA/HFSA 2016 [14], ESC 2019 [15]. ACC, American College of Cardiology; EMA, European Medicines Agency; ESC, European Society of Cardiology; FDA, US Food and Drug Administration; HRS, Heart Rhythm Society
Fifty-six of the 97 requests were related to maternal exposure during pregnancy (Fig. 1). In 41 of these cases, follow-up on pregnancy outcome could be completed: 38 pregnancies were prospectively ascertained and three were retrospective reports.
Evaluation of prospectively ascertained pregnancies
Maternal characteristics
The median age of pregnant women was 29.5 years and the median BMI 22.5. Maternal characteristics are presented in detail in Table 1. The mean time at first contact was gestational week (GW) 7. Only 16 of the 38 prospectively ascertained pregnancies were actually planned (42%) and only 7 women (18%) started folic acid before conception. Contraceptive methods, mostly oral contraceptives (n = 4), were still used by 6 women at conception.Table 1 Maternal characteristics and pregnancy outcomes of ivabradine exposed pregnancies. Data are n (%) except for age, BMI, gestational week at birth, neonatal weight, length and head circumference, which are median and interquartile range
Maternal characteristics N (%) or median (IQR)
Maternal characteristics Age (n=38) 29.5 (26.5-34)
BMI (n=34) 22.5 (19.4-25.6)
Smoking (n=37)
>5 cigarettes/day 7 (19%)
<=5 cigarettes/day 4 (11%)
Alcohol (n=36)
<=1 drink/day 2 (6%)
>1 drink/day 1 (3%)
GW at first contact (n=38) 7 (5.6-11.1)
Previous pregnancies (n=38)
0 17(45%)
1 8 (21%)
2 13 (34%)
Pregnancy outcome and neonatal characteristics N (%) or median (IQR)
Pregnancy outcome (n=38) Spontaneous abortion 3
ETOP 3
Live birth 32
Pregnancy complications (n=32, only live births) Gestational diabetes 5 (16%)
Pre-eclampsia 2 (6%)
Cesarean section 15 (46%)
Preterm birth (<GW 37) 3 (9%)
Neonatal characteristics (n=33, including one pair of twins) GW at birth (n=32) 39 (37.9-39.9)
Weight, g (n=32) 3300 (2935-3610)
Length, cm (n=31) 50 (49-52.5)
HC, cm (n=30) 35 (34-36)
SGA (n=32) 2 (6%)
BMI, Body Mass Index; ETOP, elective termination of pregnancy; GW, gestational week; HC, head circumference; IQR, interquartile range; SGA, small for gestational age (<10th percentile)
Ivabradine exposure and co-medication
The most prevalent indication for ivabradine treatment was supraventricular tachycardia (n = 29) followed by heart failure (n = 3) and coronary artery disease (n = 2). Other reported indications (n = 4) included QT prolongation and sick-sinus-syndrome, although both are regarded as contraindications for ivabradine. The median daily ivabradine dose was 7.5 mg (IQR 5–10, min–max 2.5–15, n = 33).
Nearly all women were treated with ivabradine during the first trimester (n = 37) and 32 stopped treatment after recognition of pregnancy (Fig. 3), at median gestational week 6.1 (interquartile range 5.4–7.8). Cardiac decompensation or complications were not described after cessation of treatment. Only 5 women continued ivabradine throughout pregnancy or until shortly before delivery. One woman started treatment after first trimester (GW 18) of her twin pregnancy due to symptomatic tachycardia under beta-blocker therapy (Iv34 in Fig. 3).Fig. 3 Ivabradine exposure and pregnancy outcome of prospectively (n = 38) and retrospectively (n = 3) ascertained pregnancies
Cardiovascular co-medication consisted of beta-blockers (mainly metoprolol) in 15 women, mostly concomitantly (n = 7) or in replacement of ivabradine (n = 6). Two women initiated ivabradine after discontinuation of beta-blockers (Iv30 and Iv34 in Fig. 3). Six women received other cardiovascular medication, three methyldopa or amlodipine for hypertension, two digitoxin or verapamil for tachyarrhythmia, and one sildenafil and eplerenone for cardiomyopathy with cardiac failure (Iv17).
Psychiatric co-medication was reported in 7/38 women (5 of these with an anxiety disorder) including the teratogenic antiepileptic topiramate (n = 1) and pregabalin (n = 3).
Pregnancy outcome
Of the 38 prospectively ascertained ivabradine-exposed pregnancies, 32 were live births (33 infants including one pair of twins) and three resulted in early spontaneous abortions (Iv10, Iv17, and Iv19 in Fig. 3). One of these occurred in a woman with cardiomyopathy (Iv17) and various co-medications including hydroxycarbamide for thrombocythemia until GW 7. Three pregnancies were electively terminated for social/psychological reasons. Further information on pregnancy complications and outcomes is given in Table 1 and Fig. 3.
Major birth defects were reported in one pregnancy (Iv24) with 15 mg ivabradine plus bisoprolol until GW 7 + 5, both replaced by metoprolol. Polyhydramnios and preeclampsia led to Caesarean section in GW 35 after premature rupture of membranes. The premature child was large for gestational age (> 97 percentile) and had a cleft palate, an atrial septal defect II (spontaneously closed at the age of 10 months), and a hypertrophic cardiac septum. The mother was treated for sinus tachycardia, in addition she had diabetes type I and a BMI of 30.
Neonatal characteristics
Characteristics of the 33 neonates are summarized in Table 1. Compared to the German perinatal survey, standardized birth weights were lower with a median SDS of − 0.36, corresponding to a reduction of birth weight of 149 g for singletons born in gestational week 40.
Two neonates were small for gestational age, one of the twins (exposure only during 2nd and 3rd trimester, Iv34) and one after maternal ivabradine exposure until GW 12 (Iv29). Two neonates were large for gestational age, both mothers were diabetic (Iv24, type I diabetes and Iv27, gestational diabetes).
Retrospective cases
During the study period three pregnancies were retrospectively reported (further details are given in Table 2 and Fig. 3). One premature neonate (IvR1) with tracheal atresia died perinatally. The mother had been treated with ivabradine, naproxen, and metamizole during the first trimester until recognition of pregnancy (no further details available).Table 2 Retrospectively reported pregnancies
Case Maternal age/BMI Treatment indication Ivabradine exposure, other co-medication (GW or trimester) Pregnancy outcome, GW, birth weight percentile, sex Pregnancy complications, birth defects, and additional findings
IvR1 31 years/31.2 Tachycardia Ivabradine (1st trim)
Naproxen (1st trim)
Metamizole (1st trim)
Live birth, 36 + 2, > 97th, f Polyhydramnios, perinatal death, tracheal atresia, normal karyotype (46, XX)
IvR2 29 years/18.4 Tachycardia, syncopes, pacemaker Ivabradine (1st trim)
Bisoprolol (0–35 + 5)
Levetiracetam (0–35 + 5, grand mal epilepsy)
Live birth, 35 + 5, < 10th, f IUGR (since GW 30), oligohydramnios (since GW 34), Caesarean section (vaginal bleeding)
IvR3 29 years/not available Tachycardia Ivabradine (0–41 + 3)
Bisoprolol (0–41 + 3)
Live birth, 41 + 3
> 10th, m
Uncomplicated pregnancy
BMI, body mass index; f, female; GW, gestational week; IUGR, intrauterine growth restriction; m, male; trim, trimester
The other two retrospective cases include a woman with comorbidities, intrauterine growth restriction, and premature birth (IvR2). The second was a healthy newborn (IvR3) after treatment with ivabradine throughout pregnancy. This mother reported a previous also uneventful pregnancy course under ivabradine.
Discussion
Pregnancy outcome
In our prospective cohort, there was only one neonate with major birth defects (cleft palate and atrial septal defect) from a diabetic mother with supraventricular tachycardia. No further major or cardiac malformations were reported among the 32 live births and 6 pregnancy losses. Cardiac defects have been observed in preclinical experimental studies. As a possible teratogenic mechanism, ivabradine-induced embryonic heart rate reduction leading to reduced cardiac output and hypoxia was discussed [16]. With a prevalence of almost 1 of 100 cardiac malformations belong to the most frequent birth defects [17, 18]. In our case, the neonate was large for gestational age and had a septum hypertrophy, supporting the assumption that maternal pre-gestational diabetes may also be causative for the observed cardiac defect and cleft palate. A causal relationship between ivabradine and cleft palate is unlikely, because ivabradine was discontinued at GW 7 + 5 before the beginning of the critical time period for the development of cleft palate at gestational week 9. The rate of major birth defects in our prospective cohort is comparable to the expected background rate of 2–3% [17, 18]. However, the limited number of pregnancies exposed during the entire period of embryogenesis does not allow excluding an increased risk, so far.
There was only one retrospectively reported case with a birth defect (malformation of the trachea) suspected as potentially ivabradine related. The concomitant exposures with naproxen and metamizole have not been evidenced as teratogens in human, whereas the high maternal BMI in the present case poses a risk for birth defects.
Taken together, we did not observe a specific pattern of congenital anomalies in our study. In addition, the presence of maternal comorbidities like diabetes and high BMI may have contributed to the observed birth defects.
The number of spontaneous abortions (3/38) was not higher than expected. The rate of preeclampsia and preterm birth in our cohort was within the expected range [19].
Treatment indications
Although the approved treatment indications for ivabradine, chronic heart failure and chronic stable angina pectoris (see Fig. 2 for changes of licenced use during the study period), are rare in women of reproductive age, we have observed an increasing need for information for off-label use in (younger) women considering pregnancy.
In our case series, most women were treated (off-label) for supraventricular tachycardia using lower doses than recommended (5–7.5 mg twice daily). Supraventricular tachycardia is not uncommon, especially in younger women, and is expected in 0.02–0.5% of pregnancies [20]. Aggravation of symptoms during pregnancy is described in about 20–50% of women related to various mechanisms, including pregnancy-related increase in cardiac heart rate and output and hormonal changes [20].
In 2015, the HRS (Heart Rhythm Society) considered ivabradine as promising for treating patients with inappropriate sinus tachycardia (IST) [13], followed by a recommendation for ivabradine by the ACC (American College of Cardiology) in 2016 [14]—though only one small randomized controlled study has yet supported beneficial effects [21]. Although still not approved for this indication, recent ESC guidelines [15] also recommend ivabradine alone or in combination with a beta-blocker in symptomatic patients with IST, for the treatment of focal atrial tachycardia or for postural orthostatic tachycardia syndrome (POTS), a condition considered as common cause for orthostatic intolerance in women between 15 and 25 years [20, 22] (Fig. 2).
Beta-blockers were required in 18% of women in our prospective cohort (Fig. 3). Discontinuation of ivabradine or switching to beta-blockers after recognition of pregnancy was well tolerated in most of our patients. Thus, necessity of ivabradine treatment should be critically considered in women with supraventricular tachycardia planning pregnancy. Data from previous studies suggest that discontinuation of ivabradine is possible in many (non-pregnant) patients with IST without recurrence. Even beneficial long-term effects after cessation of therapy are discussed [23, 24].
A high number of women in our cohort received therapy for anxiety and/or other psychiatric conditions. This is in line with findings from other studies reporting high rates of anxiety (25%) and depression (26%) in IST patients [25]. Anxiety is considered as non-cardiac symptom of inappropriate sinus tachycardia, and vice versa, anxiety may trigger tachycardia. Especially during pregnancy, empathic care and reassurance are necessary, to prevent increased concerns about possible harm to the fetus [26].
Preconception counselling is of utmost importance in all women considering pregnancy under medication [27]. Sinus tachycardia is a condition diagnosed mainly in young females and the first diagnosis is frequently made during or after pregnancy [25]. Considering the increasing off-label use of ivabradine as well as the high rate of > 40% unplanned pregnancies [28], more exposed pregnancies can be expected in the future. Well-established treatment options such as selected beta-blockers with low reproductive risk [29] should be considered in women of childbearing age before prescription of ivabradine.
Strength and limitations
A major strength of our study is the detailed exposure protocol and the multi-source ascertainment of pregnancy outcome via patients and their HCP followed by a case by case plausibility test.
Our data may not be representative of the German pregnant population; insofar, women with higher education are over-represented [30]. However, the regional distribution of enrolled patients is representative of the female population of childbearing age in Germany [11]. In addition, women on long-term ivabradine with severe cardiovascular conditions such as the approved treatment indications coronary artery disease and cardiac failure are under-represented in our cohort possibly favouring uneventful pregnancy outcomes. However, given the rarity of severe cardiac conditions in pregnant women, large study cohorts are not to be expected in the foreseeable future.
Conclusions
Our findings do not indicate that ivabradine is a major teratogen when used in early pregnancy. Although being the largest study published so far on the outcome of ivabradine-exposed pregnancies, the sample is too small to rule out embryotoxic effects. Further studies are needed to confirm or refute our findings.
In women planning pregnancy, established drugs of choice with low risk for the unborn should be preferred to ivabradine. For women with significant arrhythmias and severe cardiac disease, the impact of the underlying medical condition including comorbidities such as diabetes and high BMI on pregnancy outcome has to be considered when planning pregnancy.
After inadvertent ivabradine exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, all pregnant women with complicated arrhythmias or major cardiac disease should be carefully monitored.
Acknowledgements
We would like to thank all patients and physicians reporting pregnancies to the institute and acknowledge our colleagues from the German Embryotox Institute for counselling and thoroughly documenting cases, especially Sandra Zinke for continuous support with case documentation.
Contributions of authors
MH, ML, and CS designed the study. ML, EB, MH, and KD validated and analysed the data. ML, MH, and CS wrote the first draft of the manuscript, and all the authors critically revised subsequent manuscript drafts and contributed essential discussion points. All the authors approved the final version of this manuscript and they are responsible for the accuracy of this work.
Funding
Open Access funding enabled and organized by Projekt DEAL. This work was performed with financial support from the German Ministry of Health (BMG) and the German Federal Institute for Drugs and Medical Devices (BfArM). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
The study was registered with the German Clinical Trial register (DRKS00010502) and the study protocol was approved by the ethics committee of the Charité-Universitätsmedizin Berlin (EA1/107/16).
The study was registered with the German Clinical Trial register (DRKS00010502, 20.5.2016)
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | BISOPROLOL, IVABRADINE, METOPROLOL | DrugsGivenReaction | CC BY | 33501507 | 19,157,674 | 2021-07 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Premature delivery'. | Ivabradine use in pregnant women-treatment indications and pregnancy outcome: an evaluation of the German Embryotox database.
OBJECTIVE
Ivabradine has been approved for the treatment of chronic heart failure and chronic stable angina pectoris in Europe. Based on adverse outcomes of reproductive animal studies and the lack of human data, ivabradine is considered contraindicated during pregnancy. The aim of this observational study is to analyse ivabradine use before and during pregnancy.
METHODS
We evaluated all ivabradine-related requests to the German Embryotox Institute from 2007 to 2019. Exposed pregnancies were analysed as to their outcome.
RESULTS
Off-label use for supraventricular tachycardia was frequent in women of childbearing age. Of 38 prospectively ascertained pregnancies with ivabradine exposure and completed follow-up, 32 resulted in live births, 3 in spontaneous abortions, and 3 were electively terminated. One neonate presented with major birth defects (atrial septal defect and cleft palate). In 33/38 patients, ivabradine was discontinued after confirmation of pregnancy without cardiac deterioration and 5/38 women continued ivabradine throughout pregnancy. In addition, there were 3 retrospectively reported pregnancies including one major birth defect (tracheal atresia).
CONCLUSIONS
This case series represents the largest cohort of ivabradine-exposed pregnancies, published so far. According to our findings, ivabradine appears not to be a major teratogen. However, established drugs of choice with strong evidence of low risk for the unborn should be preferred in women planning pregnancy. After inadvertent exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, close monitoring is necessary in pregnant women with supraventricular arrhythmias or cardiac disease.
Introduction
Ivabradine is a heart rate-lowering agent, acting through selective inhibition of the If (I-funny) channel without additional hemodynamic effects on cardiac function. The approved treatment indications in Europe are chronic heart failure and symptomatic treatment of chronic stable angina pectoris after failing or intolerance of beta-blockers [1].
Preclinical studies have suggested fetotoxicity and teratogenic effects. In pregnant rats, ivabradine exposure equivalent to therapeutic levels in humans was associated with increased intrauterine and postnatal mortality. Ventricular septal defects and complex anomalies of the great arteries were noted at doses 3 times the therapeutic human exposure. In pregnant rabbits, reduced fetal and placental weight as well as ectrodactyly was observed at doses 15 times the therapeutic human exposure [2]. In addition, dose-dependent mortality was seen in chicken and mice embryos [3, 4].
ESC guidelines consider ivabradine contraindicated during pregnancy and recommend discontinuation independent of treatment indication before conception with close clinical and echocardiographic monitoring [5].
Published data on ivabradine-exposed pregnancies are limited to three case reports, one with ivabradine treatment in the late first and early second trimester for sinus tachycardia after myocardial infarction [6], and the second starting ivabradine in the second trimester for deteriorating tachycardia-induced cardiomyopathy [7]. In the third case with first trimester exposure, pregnancy was electively terminated [8].
The limited human experience along with the teratogenic effects in animal studies makes it difficult to counsel inadvertently exposed pregnant women and their health care providers (HCP). Therefore, we decided to evaluate ivabradine-exposed pregnancies recorded by the German Embryotox Institute.
Methods
Study design
The German Embryotox Institute offers risk assessment on drug exposure during pregnancy to HCPs and patients. Up to 15,000 annual requests are answered, and in approximately 3500 critically exposed cases per year, the course and outcome of pregnancy are documented. At the initial contact, maternal characteristics and detailed exposure assessment are asked for after informed consent. About 8 weeks after the estimated date of birth, information on course and outcome of pregnancy is collected via structured telephone interview and mailed questionnaires. After a case by case plausibility check, additional information and health care records are requested in cases of adverse outcomes or inconsistent information.
Weeks of pregnancy were based on ultrasound and if not available on the last menstrual period (LMP). First trimester was defined as gestational week 2 + 0 (conception) until 12 + 6. Major birth defects were classified according to EUROCAT [9]. In prospectively ascertained pregnancies, neither the outcome of pregnancy nor prenatal pathology was known at the time of first contact. Pregnancies that were reported after birth or because of prenatal pathology were considered retrospective and evaluated separately. A detailed description of methodology adapted to the recommendations of the Strengthening of the Reporting of Observational studies in Epidemiology (STROBE) statement is given in Schaefer et al. [10] and Dathe and Schaefer [11].
We evaluated all requests on ivabradine to the German Embryotox Institute from 2007 to 2019.
Statistical analyses
Descriptive statistics were applied. For age, BMI, gestational weeks, neonatal weight, length, and head circumference, median and interquartile ranges are presented. Birth weights were adjusted to sex and gestational age at birth and percentile categories and standard deviation scores (SDS) were calculated according to the German perinatal survey [12].
Results
During the study period 2007–2019, our institute received 97 requests for ivabradine, 56 (58%) were from HCP, and 41 from patients. For further details, see Fig. 1. Treatment indications for ivabradine over time in relation to formal approval status and pregnancy labelling are summarized in Fig. 2. Supraventricular tachycardia was the most common reason for ivabradine treatment in our cohort. The first ivabradine-exposed pregnancy was recorded in March 2007, 1 year after approval and marketing authorization in Germany in January 2006.Fig. 1 Overview on information requests to the German Embryotox institute on ivabradine
Fig. 2 Treatment indication for ivabradine in 97 Embryotox requests over time in relation to pregnancy labelling and approval status by EMA/FDA. Recommendations for use of ivabradine in supraventricular tachycardias, i.e. inappropriate sinus tachycardia and postural orthostatic tachycardia syndrome: HRS/ESC 2015 [13], ACC/AHA/HFSA 2016 [14], ESC 2019 [15]. ACC, American College of Cardiology; EMA, European Medicines Agency; ESC, European Society of Cardiology; FDA, US Food and Drug Administration; HRS, Heart Rhythm Society
Fifty-six of the 97 requests were related to maternal exposure during pregnancy (Fig. 1). In 41 of these cases, follow-up on pregnancy outcome could be completed: 38 pregnancies were prospectively ascertained and three were retrospective reports.
Evaluation of prospectively ascertained pregnancies
Maternal characteristics
The median age of pregnant women was 29.5 years and the median BMI 22.5. Maternal characteristics are presented in detail in Table 1. The mean time at first contact was gestational week (GW) 7. Only 16 of the 38 prospectively ascertained pregnancies were actually planned (42%) and only 7 women (18%) started folic acid before conception. Contraceptive methods, mostly oral contraceptives (n = 4), were still used by 6 women at conception.Table 1 Maternal characteristics and pregnancy outcomes of ivabradine exposed pregnancies. Data are n (%) except for age, BMI, gestational week at birth, neonatal weight, length and head circumference, which are median and interquartile range
Maternal characteristics N (%) or median (IQR)
Maternal characteristics Age (n=38) 29.5 (26.5-34)
BMI (n=34) 22.5 (19.4-25.6)
Smoking (n=37)
>5 cigarettes/day 7 (19%)
<=5 cigarettes/day 4 (11%)
Alcohol (n=36)
<=1 drink/day 2 (6%)
>1 drink/day 1 (3%)
GW at first contact (n=38) 7 (5.6-11.1)
Previous pregnancies (n=38)
0 17(45%)
1 8 (21%)
2 13 (34%)
Pregnancy outcome and neonatal characteristics N (%) or median (IQR)
Pregnancy outcome (n=38) Spontaneous abortion 3
ETOP 3
Live birth 32
Pregnancy complications (n=32, only live births) Gestational diabetes 5 (16%)
Pre-eclampsia 2 (6%)
Cesarean section 15 (46%)
Preterm birth (<GW 37) 3 (9%)
Neonatal characteristics (n=33, including one pair of twins) GW at birth (n=32) 39 (37.9-39.9)
Weight, g (n=32) 3300 (2935-3610)
Length, cm (n=31) 50 (49-52.5)
HC, cm (n=30) 35 (34-36)
SGA (n=32) 2 (6%)
BMI, Body Mass Index; ETOP, elective termination of pregnancy; GW, gestational week; HC, head circumference; IQR, interquartile range; SGA, small for gestational age (<10th percentile)
Ivabradine exposure and co-medication
The most prevalent indication for ivabradine treatment was supraventricular tachycardia (n = 29) followed by heart failure (n = 3) and coronary artery disease (n = 2). Other reported indications (n = 4) included QT prolongation and sick-sinus-syndrome, although both are regarded as contraindications for ivabradine. The median daily ivabradine dose was 7.5 mg (IQR 5–10, min–max 2.5–15, n = 33).
Nearly all women were treated with ivabradine during the first trimester (n = 37) and 32 stopped treatment after recognition of pregnancy (Fig. 3), at median gestational week 6.1 (interquartile range 5.4–7.8). Cardiac decompensation or complications were not described after cessation of treatment. Only 5 women continued ivabradine throughout pregnancy or until shortly before delivery. One woman started treatment after first trimester (GW 18) of her twin pregnancy due to symptomatic tachycardia under beta-blocker therapy (Iv34 in Fig. 3).Fig. 3 Ivabradine exposure and pregnancy outcome of prospectively (n = 38) and retrospectively (n = 3) ascertained pregnancies
Cardiovascular co-medication consisted of beta-blockers (mainly metoprolol) in 15 women, mostly concomitantly (n = 7) or in replacement of ivabradine (n = 6). Two women initiated ivabradine after discontinuation of beta-blockers (Iv30 and Iv34 in Fig. 3). Six women received other cardiovascular medication, three methyldopa or amlodipine for hypertension, two digitoxin or verapamil for tachyarrhythmia, and one sildenafil and eplerenone for cardiomyopathy with cardiac failure (Iv17).
Psychiatric co-medication was reported in 7/38 women (5 of these with an anxiety disorder) including the teratogenic antiepileptic topiramate (n = 1) and pregabalin (n = 3).
Pregnancy outcome
Of the 38 prospectively ascertained ivabradine-exposed pregnancies, 32 were live births (33 infants including one pair of twins) and three resulted in early spontaneous abortions (Iv10, Iv17, and Iv19 in Fig. 3). One of these occurred in a woman with cardiomyopathy (Iv17) and various co-medications including hydroxycarbamide for thrombocythemia until GW 7. Three pregnancies were electively terminated for social/psychological reasons. Further information on pregnancy complications and outcomes is given in Table 1 and Fig. 3.
Major birth defects were reported in one pregnancy (Iv24) with 15 mg ivabradine plus bisoprolol until GW 7 + 5, both replaced by metoprolol. Polyhydramnios and preeclampsia led to Caesarean section in GW 35 after premature rupture of membranes. The premature child was large for gestational age (> 97 percentile) and had a cleft palate, an atrial septal defect II (spontaneously closed at the age of 10 months), and a hypertrophic cardiac septum. The mother was treated for sinus tachycardia, in addition she had diabetes type I and a BMI of 30.
Neonatal characteristics
Characteristics of the 33 neonates are summarized in Table 1. Compared to the German perinatal survey, standardized birth weights were lower with a median SDS of − 0.36, corresponding to a reduction of birth weight of 149 g for singletons born in gestational week 40.
Two neonates were small for gestational age, one of the twins (exposure only during 2nd and 3rd trimester, Iv34) and one after maternal ivabradine exposure until GW 12 (Iv29). Two neonates were large for gestational age, both mothers were diabetic (Iv24, type I diabetes and Iv27, gestational diabetes).
Retrospective cases
During the study period three pregnancies were retrospectively reported (further details are given in Table 2 and Fig. 3). One premature neonate (IvR1) with tracheal atresia died perinatally. The mother had been treated with ivabradine, naproxen, and metamizole during the first trimester until recognition of pregnancy (no further details available).Table 2 Retrospectively reported pregnancies
Case Maternal age/BMI Treatment indication Ivabradine exposure, other co-medication (GW or trimester) Pregnancy outcome, GW, birth weight percentile, sex Pregnancy complications, birth defects, and additional findings
IvR1 31 years/31.2 Tachycardia Ivabradine (1st trim)
Naproxen (1st trim)
Metamizole (1st trim)
Live birth, 36 + 2, > 97th, f Polyhydramnios, perinatal death, tracheal atresia, normal karyotype (46, XX)
IvR2 29 years/18.4 Tachycardia, syncopes, pacemaker Ivabradine (1st trim)
Bisoprolol (0–35 + 5)
Levetiracetam (0–35 + 5, grand mal epilepsy)
Live birth, 35 + 5, < 10th, f IUGR (since GW 30), oligohydramnios (since GW 34), Caesarean section (vaginal bleeding)
IvR3 29 years/not available Tachycardia Ivabradine (0–41 + 3)
Bisoprolol (0–41 + 3)
Live birth, 41 + 3
> 10th, m
Uncomplicated pregnancy
BMI, body mass index; f, female; GW, gestational week; IUGR, intrauterine growth restriction; m, male; trim, trimester
The other two retrospective cases include a woman with comorbidities, intrauterine growth restriction, and premature birth (IvR2). The second was a healthy newborn (IvR3) after treatment with ivabradine throughout pregnancy. This mother reported a previous also uneventful pregnancy course under ivabradine.
Discussion
Pregnancy outcome
In our prospective cohort, there was only one neonate with major birth defects (cleft palate and atrial septal defect) from a diabetic mother with supraventricular tachycardia. No further major or cardiac malformations were reported among the 32 live births and 6 pregnancy losses. Cardiac defects have been observed in preclinical experimental studies. As a possible teratogenic mechanism, ivabradine-induced embryonic heart rate reduction leading to reduced cardiac output and hypoxia was discussed [16]. With a prevalence of almost 1 of 100 cardiac malformations belong to the most frequent birth defects [17, 18]. In our case, the neonate was large for gestational age and had a septum hypertrophy, supporting the assumption that maternal pre-gestational diabetes may also be causative for the observed cardiac defect and cleft palate. A causal relationship between ivabradine and cleft palate is unlikely, because ivabradine was discontinued at GW 7 + 5 before the beginning of the critical time period for the development of cleft palate at gestational week 9. The rate of major birth defects in our prospective cohort is comparable to the expected background rate of 2–3% [17, 18]. However, the limited number of pregnancies exposed during the entire period of embryogenesis does not allow excluding an increased risk, so far.
There was only one retrospectively reported case with a birth defect (malformation of the trachea) suspected as potentially ivabradine related. The concomitant exposures with naproxen and metamizole have not been evidenced as teratogens in human, whereas the high maternal BMI in the present case poses a risk for birth defects.
Taken together, we did not observe a specific pattern of congenital anomalies in our study. In addition, the presence of maternal comorbidities like diabetes and high BMI may have contributed to the observed birth defects.
The number of spontaneous abortions (3/38) was not higher than expected. The rate of preeclampsia and preterm birth in our cohort was within the expected range [19].
Treatment indications
Although the approved treatment indications for ivabradine, chronic heart failure and chronic stable angina pectoris (see Fig. 2 for changes of licenced use during the study period), are rare in women of reproductive age, we have observed an increasing need for information for off-label use in (younger) women considering pregnancy.
In our case series, most women were treated (off-label) for supraventricular tachycardia using lower doses than recommended (5–7.5 mg twice daily). Supraventricular tachycardia is not uncommon, especially in younger women, and is expected in 0.02–0.5% of pregnancies [20]. Aggravation of symptoms during pregnancy is described in about 20–50% of women related to various mechanisms, including pregnancy-related increase in cardiac heart rate and output and hormonal changes [20].
In 2015, the HRS (Heart Rhythm Society) considered ivabradine as promising for treating patients with inappropriate sinus tachycardia (IST) [13], followed by a recommendation for ivabradine by the ACC (American College of Cardiology) in 2016 [14]—though only one small randomized controlled study has yet supported beneficial effects [21]. Although still not approved for this indication, recent ESC guidelines [15] also recommend ivabradine alone or in combination with a beta-blocker in symptomatic patients with IST, for the treatment of focal atrial tachycardia or for postural orthostatic tachycardia syndrome (POTS), a condition considered as common cause for orthostatic intolerance in women between 15 and 25 years [20, 22] (Fig. 2).
Beta-blockers were required in 18% of women in our prospective cohort (Fig. 3). Discontinuation of ivabradine or switching to beta-blockers after recognition of pregnancy was well tolerated in most of our patients. Thus, necessity of ivabradine treatment should be critically considered in women with supraventricular tachycardia planning pregnancy. Data from previous studies suggest that discontinuation of ivabradine is possible in many (non-pregnant) patients with IST without recurrence. Even beneficial long-term effects after cessation of therapy are discussed [23, 24].
A high number of women in our cohort received therapy for anxiety and/or other psychiatric conditions. This is in line with findings from other studies reporting high rates of anxiety (25%) and depression (26%) in IST patients [25]. Anxiety is considered as non-cardiac symptom of inappropriate sinus tachycardia, and vice versa, anxiety may trigger tachycardia. Especially during pregnancy, empathic care and reassurance are necessary, to prevent increased concerns about possible harm to the fetus [26].
Preconception counselling is of utmost importance in all women considering pregnancy under medication [27]. Sinus tachycardia is a condition diagnosed mainly in young females and the first diagnosis is frequently made during or after pregnancy [25]. Considering the increasing off-label use of ivabradine as well as the high rate of > 40% unplanned pregnancies [28], more exposed pregnancies can be expected in the future. Well-established treatment options such as selected beta-blockers with low reproductive risk [29] should be considered in women of childbearing age before prescription of ivabradine.
Strength and limitations
A major strength of our study is the detailed exposure protocol and the multi-source ascertainment of pregnancy outcome via patients and their HCP followed by a case by case plausibility test.
Our data may not be representative of the German pregnant population; insofar, women with higher education are over-represented [30]. However, the regional distribution of enrolled patients is representative of the female population of childbearing age in Germany [11]. In addition, women on long-term ivabradine with severe cardiovascular conditions such as the approved treatment indications coronary artery disease and cardiac failure are under-represented in our cohort possibly favouring uneventful pregnancy outcomes. However, given the rarity of severe cardiac conditions in pregnant women, large study cohorts are not to be expected in the foreseeable future.
Conclusions
Our findings do not indicate that ivabradine is a major teratogen when used in early pregnancy. Although being the largest study published so far on the outcome of ivabradine-exposed pregnancies, the sample is too small to rule out embryotoxic effects. Further studies are needed to confirm or refute our findings.
In women planning pregnancy, established drugs of choice with low risk for the unborn should be preferred to ivabradine. For women with significant arrhythmias and severe cardiac disease, the impact of the underlying medical condition including comorbidities such as diabetes and high BMI on pregnancy outcome has to be considered when planning pregnancy.
After inadvertent ivabradine exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, all pregnant women with complicated arrhythmias or major cardiac disease should be carefully monitored.
Acknowledgements
We would like to thank all patients and physicians reporting pregnancies to the institute and acknowledge our colleagues from the German Embryotox Institute for counselling and thoroughly documenting cases, especially Sandra Zinke for continuous support with case documentation.
Contributions of authors
MH, ML, and CS designed the study. ML, EB, MH, and KD validated and analysed the data. ML, MH, and CS wrote the first draft of the manuscript, and all the authors critically revised subsequent manuscript drafts and contributed essential discussion points. All the authors approved the final version of this manuscript and they are responsible for the accuracy of this work.
Funding
Open Access funding enabled and organized by Projekt DEAL. This work was performed with financial support from the German Ministry of Health (BMG) and the German Federal Institute for Drugs and Medical Devices (BfArM). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
The study was registered with the German Clinical Trial register (DRKS00010502) and the study protocol was approved by the ethics committee of the Charité-Universitätsmedizin Berlin (EA1/107/16).
The study was registered with the German Clinical Trial register (DRKS00010502, 20.5.2016)
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | BISOPROLOL, IVABRADINE, LEVETIRACETAM, METOPROLOL | DrugsGivenReaction | CC BY | 33501507 | 18,911,414 | 2021-07 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Premature rupture of membranes'. | Ivabradine use in pregnant women-treatment indications and pregnancy outcome: an evaluation of the German Embryotox database.
OBJECTIVE
Ivabradine has been approved for the treatment of chronic heart failure and chronic stable angina pectoris in Europe. Based on adverse outcomes of reproductive animal studies and the lack of human data, ivabradine is considered contraindicated during pregnancy. The aim of this observational study is to analyse ivabradine use before and during pregnancy.
METHODS
We evaluated all ivabradine-related requests to the German Embryotox Institute from 2007 to 2019. Exposed pregnancies were analysed as to their outcome.
RESULTS
Off-label use for supraventricular tachycardia was frequent in women of childbearing age. Of 38 prospectively ascertained pregnancies with ivabradine exposure and completed follow-up, 32 resulted in live births, 3 in spontaneous abortions, and 3 were electively terminated. One neonate presented with major birth defects (atrial septal defect and cleft palate). In 33/38 patients, ivabradine was discontinued after confirmation of pregnancy without cardiac deterioration and 5/38 women continued ivabradine throughout pregnancy. In addition, there were 3 retrospectively reported pregnancies including one major birth defect (tracheal atresia).
CONCLUSIONS
This case series represents the largest cohort of ivabradine-exposed pregnancies, published so far. According to our findings, ivabradine appears not to be a major teratogen. However, established drugs of choice with strong evidence of low risk for the unborn should be preferred in women planning pregnancy. After inadvertent exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, close monitoring is necessary in pregnant women with supraventricular arrhythmias or cardiac disease.
Introduction
Ivabradine is a heart rate-lowering agent, acting through selective inhibition of the If (I-funny) channel without additional hemodynamic effects on cardiac function. The approved treatment indications in Europe are chronic heart failure and symptomatic treatment of chronic stable angina pectoris after failing or intolerance of beta-blockers [1].
Preclinical studies have suggested fetotoxicity and teratogenic effects. In pregnant rats, ivabradine exposure equivalent to therapeutic levels in humans was associated with increased intrauterine and postnatal mortality. Ventricular septal defects and complex anomalies of the great arteries were noted at doses 3 times the therapeutic human exposure. In pregnant rabbits, reduced fetal and placental weight as well as ectrodactyly was observed at doses 15 times the therapeutic human exposure [2]. In addition, dose-dependent mortality was seen in chicken and mice embryos [3, 4].
ESC guidelines consider ivabradine contraindicated during pregnancy and recommend discontinuation independent of treatment indication before conception with close clinical and echocardiographic monitoring [5].
Published data on ivabradine-exposed pregnancies are limited to three case reports, one with ivabradine treatment in the late first and early second trimester for sinus tachycardia after myocardial infarction [6], and the second starting ivabradine in the second trimester for deteriorating tachycardia-induced cardiomyopathy [7]. In the third case with first trimester exposure, pregnancy was electively terminated [8].
The limited human experience along with the teratogenic effects in animal studies makes it difficult to counsel inadvertently exposed pregnant women and their health care providers (HCP). Therefore, we decided to evaluate ivabradine-exposed pregnancies recorded by the German Embryotox Institute.
Methods
Study design
The German Embryotox Institute offers risk assessment on drug exposure during pregnancy to HCPs and patients. Up to 15,000 annual requests are answered, and in approximately 3500 critically exposed cases per year, the course and outcome of pregnancy are documented. At the initial contact, maternal characteristics and detailed exposure assessment are asked for after informed consent. About 8 weeks after the estimated date of birth, information on course and outcome of pregnancy is collected via structured telephone interview and mailed questionnaires. After a case by case plausibility check, additional information and health care records are requested in cases of adverse outcomes or inconsistent information.
Weeks of pregnancy were based on ultrasound and if not available on the last menstrual period (LMP). First trimester was defined as gestational week 2 + 0 (conception) until 12 + 6. Major birth defects were classified according to EUROCAT [9]. In prospectively ascertained pregnancies, neither the outcome of pregnancy nor prenatal pathology was known at the time of first contact. Pregnancies that were reported after birth or because of prenatal pathology were considered retrospective and evaluated separately. A detailed description of methodology adapted to the recommendations of the Strengthening of the Reporting of Observational studies in Epidemiology (STROBE) statement is given in Schaefer et al. [10] and Dathe and Schaefer [11].
We evaluated all requests on ivabradine to the German Embryotox Institute from 2007 to 2019.
Statistical analyses
Descriptive statistics were applied. For age, BMI, gestational weeks, neonatal weight, length, and head circumference, median and interquartile ranges are presented. Birth weights were adjusted to sex and gestational age at birth and percentile categories and standard deviation scores (SDS) were calculated according to the German perinatal survey [12].
Results
During the study period 2007–2019, our institute received 97 requests for ivabradine, 56 (58%) were from HCP, and 41 from patients. For further details, see Fig. 1. Treatment indications for ivabradine over time in relation to formal approval status and pregnancy labelling are summarized in Fig. 2. Supraventricular tachycardia was the most common reason for ivabradine treatment in our cohort. The first ivabradine-exposed pregnancy was recorded in March 2007, 1 year after approval and marketing authorization in Germany in January 2006.Fig. 1 Overview on information requests to the German Embryotox institute on ivabradine
Fig. 2 Treatment indication for ivabradine in 97 Embryotox requests over time in relation to pregnancy labelling and approval status by EMA/FDA. Recommendations for use of ivabradine in supraventricular tachycardias, i.e. inappropriate sinus tachycardia and postural orthostatic tachycardia syndrome: HRS/ESC 2015 [13], ACC/AHA/HFSA 2016 [14], ESC 2019 [15]. ACC, American College of Cardiology; EMA, European Medicines Agency; ESC, European Society of Cardiology; FDA, US Food and Drug Administration; HRS, Heart Rhythm Society
Fifty-six of the 97 requests were related to maternal exposure during pregnancy (Fig. 1). In 41 of these cases, follow-up on pregnancy outcome could be completed: 38 pregnancies were prospectively ascertained and three were retrospective reports.
Evaluation of prospectively ascertained pregnancies
Maternal characteristics
The median age of pregnant women was 29.5 years and the median BMI 22.5. Maternal characteristics are presented in detail in Table 1. The mean time at first contact was gestational week (GW) 7. Only 16 of the 38 prospectively ascertained pregnancies were actually planned (42%) and only 7 women (18%) started folic acid before conception. Contraceptive methods, mostly oral contraceptives (n = 4), were still used by 6 women at conception.Table 1 Maternal characteristics and pregnancy outcomes of ivabradine exposed pregnancies. Data are n (%) except for age, BMI, gestational week at birth, neonatal weight, length and head circumference, which are median and interquartile range
Maternal characteristics N (%) or median (IQR)
Maternal characteristics Age (n=38) 29.5 (26.5-34)
BMI (n=34) 22.5 (19.4-25.6)
Smoking (n=37)
>5 cigarettes/day 7 (19%)
<=5 cigarettes/day 4 (11%)
Alcohol (n=36)
<=1 drink/day 2 (6%)
>1 drink/day 1 (3%)
GW at first contact (n=38) 7 (5.6-11.1)
Previous pregnancies (n=38)
0 17(45%)
1 8 (21%)
2 13 (34%)
Pregnancy outcome and neonatal characteristics N (%) or median (IQR)
Pregnancy outcome (n=38) Spontaneous abortion 3
ETOP 3
Live birth 32
Pregnancy complications (n=32, only live births) Gestational diabetes 5 (16%)
Pre-eclampsia 2 (6%)
Cesarean section 15 (46%)
Preterm birth (<GW 37) 3 (9%)
Neonatal characteristics (n=33, including one pair of twins) GW at birth (n=32) 39 (37.9-39.9)
Weight, g (n=32) 3300 (2935-3610)
Length, cm (n=31) 50 (49-52.5)
HC, cm (n=30) 35 (34-36)
SGA (n=32) 2 (6%)
BMI, Body Mass Index; ETOP, elective termination of pregnancy; GW, gestational week; HC, head circumference; IQR, interquartile range; SGA, small for gestational age (<10th percentile)
Ivabradine exposure and co-medication
The most prevalent indication for ivabradine treatment was supraventricular tachycardia (n = 29) followed by heart failure (n = 3) and coronary artery disease (n = 2). Other reported indications (n = 4) included QT prolongation and sick-sinus-syndrome, although both are regarded as contraindications for ivabradine. The median daily ivabradine dose was 7.5 mg (IQR 5–10, min–max 2.5–15, n = 33).
Nearly all women were treated with ivabradine during the first trimester (n = 37) and 32 stopped treatment after recognition of pregnancy (Fig. 3), at median gestational week 6.1 (interquartile range 5.4–7.8). Cardiac decompensation or complications were not described after cessation of treatment. Only 5 women continued ivabradine throughout pregnancy or until shortly before delivery. One woman started treatment after first trimester (GW 18) of her twin pregnancy due to symptomatic tachycardia under beta-blocker therapy (Iv34 in Fig. 3).Fig. 3 Ivabradine exposure and pregnancy outcome of prospectively (n = 38) and retrospectively (n = 3) ascertained pregnancies
Cardiovascular co-medication consisted of beta-blockers (mainly metoprolol) in 15 women, mostly concomitantly (n = 7) or in replacement of ivabradine (n = 6). Two women initiated ivabradine after discontinuation of beta-blockers (Iv30 and Iv34 in Fig. 3). Six women received other cardiovascular medication, three methyldopa or amlodipine for hypertension, two digitoxin or verapamil for tachyarrhythmia, and one sildenafil and eplerenone for cardiomyopathy with cardiac failure (Iv17).
Psychiatric co-medication was reported in 7/38 women (5 of these with an anxiety disorder) including the teratogenic antiepileptic topiramate (n = 1) and pregabalin (n = 3).
Pregnancy outcome
Of the 38 prospectively ascertained ivabradine-exposed pregnancies, 32 were live births (33 infants including one pair of twins) and three resulted in early spontaneous abortions (Iv10, Iv17, and Iv19 in Fig. 3). One of these occurred in a woman with cardiomyopathy (Iv17) and various co-medications including hydroxycarbamide for thrombocythemia until GW 7. Three pregnancies were electively terminated for social/psychological reasons. Further information on pregnancy complications and outcomes is given in Table 1 and Fig. 3.
Major birth defects were reported in one pregnancy (Iv24) with 15 mg ivabradine plus bisoprolol until GW 7 + 5, both replaced by metoprolol. Polyhydramnios and preeclampsia led to Caesarean section in GW 35 after premature rupture of membranes. The premature child was large for gestational age (> 97 percentile) and had a cleft palate, an atrial septal defect II (spontaneously closed at the age of 10 months), and a hypertrophic cardiac septum. The mother was treated for sinus tachycardia, in addition she had diabetes type I and a BMI of 30.
Neonatal characteristics
Characteristics of the 33 neonates are summarized in Table 1. Compared to the German perinatal survey, standardized birth weights were lower with a median SDS of − 0.36, corresponding to a reduction of birth weight of 149 g for singletons born in gestational week 40.
Two neonates were small for gestational age, one of the twins (exposure only during 2nd and 3rd trimester, Iv34) and one after maternal ivabradine exposure until GW 12 (Iv29). Two neonates were large for gestational age, both mothers were diabetic (Iv24, type I diabetes and Iv27, gestational diabetes).
Retrospective cases
During the study period three pregnancies were retrospectively reported (further details are given in Table 2 and Fig. 3). One premature neonate (IvR1) with tracheal atresia died perinatally. The mother had been treated with ivabradine, naproxen, and metamizole during the first trimester until recognition of pregnancy (no further details available).Table 2 Retrospectively reported pregnancies
Case Maternal age/BMI Treatment indication Ivabradine exposure, other co-medication (GW or trimester) Pregnancy outcome, GW, birth weight percentile, sex Pregnancy complications, birth defects, and additional findings
IvR1 31 years/31.2 Tachycardia Ivabradine (1st trim)
Naproxen (1st trim)
Metamizole (1st trim)
Live birth, 36 + 2, > 97th, f Polyhydramnios, perinatal death, tracheal atresia, normal karyotype (46, XX)
IvR2 29 years/18.4 Tachycardia, syncopes, pacemaker Ivabradine (1st trim)
Bisoprolol (0–35 + 5)
Levetiracetam (0–35 + 5, grand mal epilepsy)
Live birth, 35 + 5, < 10th, f IUGR (since GW 30), oligohydramnios (since GW 34), Caesarean section (vaginal bleeding)
IvR3 29 years/not available Tachycardia Ivabradine (0–41 + 3)
Bisoprolol (0–41 + 3)
Live birth, 41 + 3
> 10th, m
Uncomplicated pregnancy
BMI, body mass index; f, female; GW, gestational week; IUGR, intrauterine growth restriction; m, male; trim, trimester
The other two retrospective cases include a woman with comorbidities, intrauterine growth restriction, and premature birth (IvR2). The second was a healthy newborn (IvR3) after treatment with ivabradine throughout pregnancy. This mother reported a previous also uneventful pregnancy course under ivabradine.
Discussion
Pregnancy outcome
In our prospective cohort, there was only one neonate with major birth defects (cleft palate and atrial septal defect) from a diabetic mother with supraventricular tachycardia. No further major or cardiac malformations were reported among the 32 live births and 6 pregnancy losses. Cardiac defects have been observed in preclinical experimental studies. As a possible teratogenic mechanism, ivabradine-induced embryonic heart rate reduction leading to reduced cardiac output and hypoxia was discussed [16]. With a prevalence of almost 1 of 100 cardiac malformations belong to the most frequent birth defects [17, 18]. In our case, the neonate was large for gestational age and had a septum hypertrophy, supporting the assumption that maternal pre-gestational diabetes may also be causative for the observed cardiac defect and cleft palate. A causal relationship between ivabradine and cleft palate is unlikely, because ivabradine was discontinued at GW 7 + 5 before the beginning of the critical time period for the development of cleft palate at gestational week 9. The rate of major birth defects in our prospective cohort is comparable to the expected background rate of 2–3% [17, 18]. However, the limited number of pregnancies exposed during the entire period of embryogenesis does not allow excluding an increased risk, so far.
There was only one retrospectively reported case with a birth defect (malformation of the trachea) suspected as potentially ivabradine related. The concomitant exposures with naproxen and metamizole have not been evidenced as teratogens in human, whereas the high maternal BMI in the present case poses a risk for birth defects.
Taken together, we did not observe a specific pattern of congenital anomalies in our study. In addition, the presence of maternal comorbidities like diabetes and high BMI may have contributed to the observed birth defects.
The number of spontaneous abortions (3/38) was not higher than expected. The rate of preeclampsia and preterm birth in our cohort was within the expected range [19].
Treatment indications
Although the approved treatment indications for ivabradine, chronic heart failure and chronic stable angina pectoris (see Fig. 2 for changes of licenced use during the study period), are rare in women of reproductive age, we have observed an increasing need for information for off-label use in (younger) women considering pregnancy.
In our case series, most women were treated (off-label) for supraventricular tachycardia using lower doses than recommended (5–7.5 mg twice daily). Supraventricular tachycardia is not uncommon, especially in younger women, and is expected in 0.02–0.5% of pregnancies [20]. Aggravation of symptoms during pregnancy is described in about 20–50% of women related to various mechanisms, including pregnancy-related increase in cardiac heart rate and output and hormonal changes [20].
In 2015, the HRS (Heart Rhythm Society) considered ivabradine as promising for treating patients with inappropriate sinus tachycardia (IST) [13], followed by a recommendation for ivabradine by the ACC (American College of Cardiology) in 2016 [14]—though only one small randomized controlled study has yet supported beneficial effects [21]. Although still not approved for this indication, recent ESC guidelines [15] also recommend ivabradine alone or in combination with a beta-blocker in symptomatic patients with IST, for the treatment of focal atrial tachycardia or for postural orthostatic tachycardia syndrome (POTS), a condition considered as common cause for orthostatic intolerance in women between 15 and 25 years [20, 22] (Fig. 2).
Beta-blockers were required in 18% of women in our prospective cohort (Fig. 3). Discontinuation of ivabradine or switching to beta-blockers after recognition of pregnancy was well tolerated in most of our patients. Thus, necessity of ivabradine treatment should be critically considered in women with supraventricular tachycardia planning pregnancy. Data from previous studies suggest that discontinuation of ivabradine is possible in many (non-pregnant) patients with IST without recurrence. Even beneficial long-term effects after cessation of therapy are discussed [23, 24].
A high number of women in our cohort received therapy for anxiety and/or other psychiatric conditions. This is in line with findings from other studies reporting high rates of anxiety (25%) and depression (26%) in IST patients [25]. Anxiety is considered as non-cardiac symptom of inappropriate sinus tachycardia, and vice versa, anxiety may trigger tachycardia. Especially during pregnancy, empathic care and reassurance are necessary, to prevent increased concerns about possible harm to the fetus [26].
Preconception counselling is of utmost importance in all women considering pregnancy under medication [27]. Sinus tachycardia is a condition diagnosed mainly in young females and the first diagnosis is frequently made during or after pregnancy [25]. Considering the increasing off-label use of ivabradine as well as the high rate of > 40% unplanned pregnancies [28], more exposed pregnancies can be expected in the future. Well-established treatment options such as selected beta-blockers with low reproductive risk [29] should be considered in women of childbearing age before prescription of ivabradine.
Strength and limitations
A major strength of our study is the detailed exposure protocol and the multi-source ascertainment of pregnancy outcome via patients and their HCP followed by a case by case plausibility test.
Our data may not be representative of the German pregnant population; insofar, women with higher education are over-represented [30]. However, the regional distribution of enrolled patients is representative of the female population of childbearing age in Germany [11]. In addition, women on long-term ivabradine with severe cardiovascular conditions such as the approved treatment indications coronary artery disease and cardiac failure are under-represented in our cohort possibly favouring uneventful pregnancy outcomes. However, given the rarity of severe cardiac conditions in pregnant women, large study cohorts are not to be expected in the foreseeable future.
Conclusions
Our findings do not indicate that ivabradine is a major teratogen when used in early pregnancy. Although being the largest study published so far on the outcome of ivabradine-exposed pregnancies, the sample is too small to rule out embryotoxic effects. Further studies are needed to confirm or refute our findings.
In women planning pregnancy, established drugs of choice with low risk for the unborn should be preferred to ivabradine. For women with significant arrhythmias and severe cardiac disease, the impact of the underlying medical condition including comorbidities such as diabetes and high BMI on pregnancy outcome has to be considered when planning pregnancy.
After inadvertent ivabradine exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, all pregnant women with complicated arrhythmias or major cardiac disease should be carefully monitored.
Acknowledgements
We would like to thank all patients and physicians reporting pregnancies to the institute and acknowledge our colleagues from the German Embryotox Institute for counselling and thoroughly documenting cases, especially Sandra Zinke for continuous support with case documentation.
Contributions of authors
MH, ML, and CS designed the study. ML, EB, MH, and KD validated and analysed the data. ML, MH, and CS wrote the first draft of the manuscript, and all the authors critically revised subsequent manuscript drafts and contributed essential discussion points. All the authors approved the final version of this manuscript and they are responsible for the accuracy of this work.
Funding
Open Access funding enabled and organized by Projekt DEAL. This work was performed with financial support from the German Ministry of Health (BMG) and the German Federal Institute for Drugs and Medical Devices (BfArM). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
The study was registered with the German Clinical Trial register (DRKS00010502) and the study protocol was approved by the ethics committee of the Charité-Universitätsmedizin Berlin (EA1/107/16).
The study was registered with the German Clinical Trial register (DRKS00010502, 20.5.2016)
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | BISOPROLOL, IVABRADINE, METOPROLOL | DrugsGivenReaction | CC BY | 33501507 | 19,157,674 | 2021-07 |
What was the dosage of drug 'BISOPROLOL'? | Ivabradine use in pregnant women-treatment indications and pregnancy outcome: an evaluation of the German Embryotox database.
OBJECTIVE
Ivabradine has been approved for the treatment of chronic heart failure and chronic stable angina pectoris in Europe. Based on adverse outcomes of reproductive animal studies and the lack of human data, ivabradine is considered contraindicated during pregnancy. The aim of this observational study is to analyse ivabradine use before and during pregnancy.
METHODS
We evaluated all ivabradine-related requests to the German Embryotox Institute from 2007 to 2019. Exposed pregnancies were analysed as to their outcome.
RESULTS
Off-label use for supraventricular tachycardia was frequent in women of childbearing age. Of 38 prospectively ascertained pregnancies with ivabradine exposure and completed follow-up, 32 resulted in live births, 3 in spontaneous abortions, and 3 were electively terminated. One neonate presented with major birth defects (atrial septal defect and cleft palate). In 33/38 patients, ivabradine was discontinued after confirmation of pregnancy without cardiac deterioration and 5/38 women continued ivabradine throughout pregnancy. In addition, there were 3 retrospectively reported pregnancies including one major birth defect (tracheal atresia).
CONCLUSIONS
This case series represents the largest cohort of ivabradine-exposed pregnancies, published so far. According to our findings, ivabradine appears not to be a major teratogen. However, established drugs of choice with strong evidence of low risk for the unborn should be preferred in women planning pregnancy. After inadvertent exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, close monitoring is necessary in pregnant women with supraventricular arrhythmias or cardiac disease.
Introduction
Ivabradine is a heart rate-lowering agent, acting through selective inhibition of the If (I-funny) channel without additional hemodynamic effects on cardiac function. The approved treatment indications in Europe are chronic heart failure and symptomatic treatment of chronic stable angina pectoris after failing or intolerance of beta-blockers [1].
Preclinical studies have suggested fetotoxicity and teratogenic effects. In pregnant rats, ivabradine exposure equivalent to therapeutic levels in humans was associated with increased intrauterine and postnatal mortality. Ventricular septal defects and complex anomalies of the great arteries were noted at doses 3 times the therapeutic human exposure. In pregnant rabbits, reduced fetal and placental weight as well as ectrodactyly was observed at doses 15 times the therapeutic human exposure [2]. In addition, dose-dependent mortality was seen in chicken and mice embryos [3, 4].
ESC guidelines consider ivabradine contraindicated during pregnancy and recommend discontinuation independent of treatment indication before conception with close clinical and echocardiographic monitoring [5].
Published data on ivabradine-exposed pregnancies are limited to three case reports, one with ivabradine treatment in the late first and early second trimester for sinus tachycardia after myocardial infarction [6], and the second starting ivabradine in the second trimester for deteriorating tachycardia-induced cardiomyopathy [7]. In the third case with first trimester exposure, pregnancy was electively terminated [8].
The limited human experience along with the teratogenic effects in animal studies makes it difficult to counsel inadvertently exposed pregnant women and their health care providers (HCP). Therefore, we decided to evaluate ivabradine-exposed pregnancies recorded by the German Embryotox Institute.
Methods
Study design
The German Embryotox Institute offers risk assessment on drug exposure during pregnancy to HCPs and patients. Up to 15,000 annual requests are answered, and in approximately 3500 critically exposed cases per year, the course and outcome of pregnancy are documented. At the initial contact, maternal characteristics and detailed exposure assessment are asked for after informed consent. About 8 weeks after the estimated date of birth, information on course and outcome of pregnancy is collected via structured telephone interview and mailed questionnaires. After a case by case plausibility check, additional information and health care records are requested in cases of adverse outcomes or inconsistent information.
Weeks of pregnancy were based on ultrasound and if not available on the last menstrual period (LMP). First trimester was defined as gestational week 2 + 0 (conception) until 12 + 6. Major birth defects were classified according to EUROCAT [9]. In prospectively ascertained pregnancies, neither the outcome of pregnancy nor prenatal pathology was known at the time of first contact. Pregnancies that were reported after birth or because of prenatal pathology were considered retrospective and evaluated separately. A detailed description of methodology adapted to the recommendations of the Strengthening of the Reporting of Observational studies in Epidemiology (STROBE) statement is given in Schaefer et al. [10] and Dathe and Schaefer [11].
We evaluated all requests on ivabradine to the German Embryotox Institute from 2007 to 2019.
Statistical analyses
Descriptive statistics were applied. For age, BMI, gestational weeks, neonatal weight, length, and head circumference, median and interquartile ranges are presented. Birth weights were adjusted to sex and gestational age at birth and percentile categories and standard deviation scores (SDS) were calculated according to the German perinatal survey [12].
Results
During the study period 2007–2019, our institute received 97 requests for ivabradine, 56 (58%) were from HCP, and 41 from patients. For further details, see Fig. 1. Treatment indications for ivabradine over time in relation to formal approval status and pregnancy labelling are summarized in Fig. 2. Supraventricular tachycardia was the most common reason for ivabradine treatment in our cohort. The first ivabradine-exposed pregnancy was recorded in March 2007, 1 year after approval and marketing authorization in Germany in January 2006.Fig. 1 Overview on information requests to the German Embryotox institute on ivabradine
Fig. 2 Treatment indication for ivabradine in 97 Embryotox requests over time in relation to pregnancy labelling and approval status by EMA/FDA. Recommendations for use of ivabradine in supraventricular tachycardias, i.e. inappropriate sinus tachycardia and postural orthostatic tachycardia syndrome: HRS/ESC 2015 [13], ACC/AHA/HFSA 2016 [14], ESC 2019 [15]. ACC, American College of Cardiology; EMA, European Medicines Agency; ESC, European Society of Cardiology; FDA, US Food and Drug Administration; HRS, Heart Rhythm Society
Fifty-six of the 97 requests were related to maternal exposure during pregnancy (Fig. 1). In 41 of these cases, follow-up on pregnancy outcome could be completed: 38 pregnancies were prospectively ascertained and three were retrospective reports.
Evaluation of prospectively ascertained pregnancies
Maternal characteristics
The median age of pregnant women was 29.5 years and the median BMI 22.5. Maternal characteristics are presented in detail in Table 1. The mean time at first contact was gestational week (GW) 7. Only 16 of the 38 prospectively ascertained pregnancies were actually planned (42%) and only 7 women (18%) started folic acid before conception. Contraceptive methods, mostly oral contraceptives (n = 4), were still used by 6 women at conception.Table 1 Maternal characteristics and pregnancy outcomes of ivabradine exposed pregnancies. Data are n (%) except for age, BMI, gestational week at birth, neonatal weight, length and head circumference, which are median and interquartile range
Maternal characteristics N (%) or median (IQR)
Maternal characteristics Age (n=38) 29.5 (26.5-34)
BMI (n=34) 22.5 (19.4-25.6)
Smoking (n=37)
>5 cigarettes/day 7 (19%)
<=5 cigarettes/day 4 (11%)
Alcohol (n=36)
<=1 drink/day 2 (6%)
>1 drink/day 1 (3%)
GW at first contact (n=38) 7 (5.6-11.1)
Previous pregnancies (n=38)
0 17(45%)
1 8 (21%)
2 13 (34%)
Pregnancy outcome and neonatal characteristics N (%) or median (IQR)
Pregnancy outcome (n=38) Spontaneous abortion 3
ETOP 3
Live birth 32
Pregnancy complications (n=32, only live births) Gestational diabetes 5 (16%)
Pre-eclampsia 2 (6%)
Cesarean section 15 (46%)
Preterm birth (<GW 37) 3 (9%)
Neonatal characteristics (n=33, including one pair of twins) GW at birth (n=32) 39 (37.9-39.9)
Weight, g (n=32) 3300 (2935-3610)
Length, cm (n=31) 50 (49-52.5)
HC, cm (n=30) 35 (34-36)
SGA (n=32) 2 (6%)
BMI, Body Mass Index; ETOP, elective termination of pregnancy; GW, gestational week; HC, head circumference; IQR, interquartile range; SGA, small for gestational age (<10th percentile)
Ivabradine exposure and co-medication
The most prevalent indication for ivabradine treatment was supraventricular tachycardia (n = 29) followed by heart failure (n = 3) and coronary artery disease (n = 2). Other reported indications (n = 4) included QT prolongation and sick-sinus-syndrome, although both are regarded as contraindications for ivabradine. The median daily ivabradine dose was 7.5 mg (IQR 5–10, min–max 2.5–15, n = 33).
Nearly all women were treated with ivabradine during the first trimester (n = 37) and 32 stopped treatment after recognition of pregnancy (Fig. 3), at median gestational week 6.1 (interquartile range 5.4–7.8). Cardiac decompensation or complications were not described after cessation of treatment. Only 5 women continued ivabradine throughout pregnancy or until shortly before delivery. One woman started treatment after first trimester (GW 18) of her twin pregnancy due to symptomatic tachycardia under beta-blocker therapy (Iv34 in Fig. 3).Fig. 3 Ivabradine exposure and pregnancy outcome of prospectively (n = 38) and retrospectively (n = 3) ascertained pregnancies
Cardiovascular co-medication consisted of beta-blockers (mainly metoprolol) in 15 women, mostly concomitantly (n = 7) or in replacement of ivabradine (n = 6). Two women initiated ivabradine after discontinuation of beta-blockers (Iv30 and Iv34 in Fig. 3). Six women received other cardiovascular medication, three methyldopa or amlodipine for hypertension, two digitoxin or verapamil for tachyarrhythmia, and one sildenafil and eplerenone for cardiomyopathy with cardiac failure (Iv17).
Psychiatric co-medication was reported in 7/38 women (5 of these with an anxiety disorder) including the teratogenic antiepileptic topiramate (n = 1) and pregabalin (n = 3).
Pregnancy outcome
Of the 38 prospectively ascertained ivabradine-exposed pregnancies, 32 were live births (33 infants including one pair of twins) and three resulted in early spontaneous abortions (Iv10, Iv17, and Iv19 in Fig. 3). One of these occurred in a woman with cardiomyopathy (Iv17) and various co-medications including hydroxycarbamide for thrombocythemia until GW 7. Three pregnancies were electively terminated for social/psychological reasons. Further information on pregnancy complications and outcomes is given in Table 1 and Fig. 3.
Major birth defects were reported in one pregnancy (Iv24) with 15 mg ivabradine plus bisoprolol until GW 7 + 5, both replaced by metoprolol. Polyhydramnios and preeclampsia led to Caesarean section in GW 35 after premature rupture of membranes. The premature child was large for gestational age (> 97 percentile) and had a cleft palate, an atrial septal defect II (spontaneously closed at the age of 10 months), and a hypertrophic cardiac septum. The mother was treated for sinus tachycardia, in addition she had diabetes type I and a BMI of 30.
Neonatal characteristics
Characteristics of the 33 neonates are summarized in Table 1. Compared to the German perinatal survey, standardized birth weights were lower with a median SDS of − 0.36, corresponding to a reduction of birth weight of 149 g for singletons born in gestational week 40.
Two neonates were small for gestational age, one of the twins (exposure only during 2nd and 3rd trimester, Iv34) and one after maternal ivabradine exposure until GW 12 (Iv29). Two neonates were large for gestational age, both mothers were diabetic (Iv24, type I diabetes and Iv27, gestational diabetes).
Retrospective cases
During the study period three pregnancies were retrospectively reported (further details are given in Table 2 and Fig. 3). One premature neonate (IvR1) with tracheal atresia died perinatally. The mother had been treated with ivabradine, naproxen, and metamizole during the first trimester until recognition of pregnancy (no further details available).Table 2 Retrospectively reported pregnancies
Case Maternal age/BMI Treatment indication Ivabradine exposure, other co-medication (GW or trimester) Pregnancy outcome, GW, birth weight percentile, sex Pregnancy complications, birth defects, and additional findings
IvR1 31 years/31.2 Tachycardia Ivabradine (1st trim)
Naproxen (1st trim)
Metamizole (1st trim)
Live birth, 36 + 2, > 97th, f Polyhydramnios, perinatal death, tracheal atresia, normal karyotype (46, XX)
IvR2 29 years/18.4 Tachycardia, syncopes, pacemaker Ivabradine (1st trim)
Bisoprolol (0–35 + 5)
Levetiracetam (0–35 + 5, grand mal epilepsy)
Live birth, 35 + 5, < 10th, f IUGR (since GW 30), oligohydramnios (since GW 34), Caesarean section (vaginal bleeding)
IvR3 29 years/not available Tachycardia Ivabradine (0–41 + 3)
Bisoprolol (0–41 + 3)
Live birth, 41 + 3
> 10th, m
Uncomplicated pregnancy
BMI, body mass index; f, female; GW, gestational week; IUGR, intrauterine growth restriction; m, male; trim, trimester
The other two retrospective cases include a woman with comorbidities, intrauterine growth restriction, and premature birth (IvR2). The second was a healthy newborn (IvR3) after treatment with ivabradine throughout pregnancy. This mother reported a previous also uneventful pregnancy course under ivabradine.
Discussion
Pregnancy outcome
In our prospective cohort, there was only one neonate with major birth defects (cleft palate and atrial septal defect) from a diabetic mother with supraventricular tachycardia. No further major or cardiac malformations were reported among the 32 live births and 6 pregnancy losses. Cardiac defects have been observed in preclinical experimental studies. As a possible teratogenic mechanism, ivabradine-induced embryonic heart rate reduction leading to reduced cardiac output and hypoxia was discussed [16]. With a prevalence of almost 1 of 100 cardiac malformations belong to the most frequent birth defects [17, 18]. In our case, the neonate was large for gestational age and had a septum hypertrophy, supporting the assumption that maternal pre-gestational diabetes may also be causative for the observed cardiac defect and cleft palate. A causal relationship between ivabradine and cleft palate is unlikely, because ivabradine was discontinued at GW 7 + 5 before the beginning of the critical time period for the development of cleft palate at gestational week 9. The rate of major birth defects in our prospective cohort is comparable to the expected background rate of 2–3% [17, 18]. However, the limited number of pregnancies exposed during the entire period of embryogenesis does not allow excluding an increased risk, so far.
There was only one retrospectively reported case with a birth defect (malformation of the trachea) suspected as potentially ivabradine related. The concomitant exposures with naproxen and metamizole have not been evidenced as teratogens in human, whereas the high maternal BMI in the present case poses a risk for birth defects.
Taken together, we did not observe a specific pattern of congenital anomalies in our study. In addition, the presence of maternal comorbidities like diabetes and high BMI may have contributed to the observed birth defects.
The number of spontaneous abortions (3/38) was not higher than expected. The rate of preeclampsia and preterm birth in our cohort was within the expected range [19].
Treatment indications
Although the approved treatment indications for ivabradine, chronic heart failure and chronic stable angina pectoris (see Fig. 2 for changes of licenced use during the study period), are rare in women of reproductive age, we have observed an increasing need for information for off-label use in (younger) women considering pregnancy.
In our case series, most women were treated (off-label) for supraventricular tachycardia using lower doses than recommended (5–7.5 mg twice daily). Supraventricular tachycardia is not uncommon, especially in younger women, and is expected in 0.02–0.5% of pregnancies [20]. Aggravation of symptoms during pregnancy is described in about 20–50% of women related to various mechanisms, including pregnancy-related increase in cardiac heart rate and output and hormonal changes [20].
In 2015, the HRS (Heart Rhythm Society) considered ivabradine as promising for treating patients with inappropriate sinus tachycardia (IST) [13], followed by a recommendation for ivabradine by the ACC (American College of Cardiology) in 2016 [14]—though only one small randomized controlled study has yet supported beneficial effects [21]. Although still not approved for this indication, recent ESC guidelines [15] also recommend ivabradine alone or in combination with a beta-blocker in symptomatic patients with IST, for the treatment of focal atrial tachycardia or for postural orthostatic tachycardia syndrome (POTS), a condition considered as common cause for orthostatic intolerance in women between 15 and 25 years [20, 22] (Fig. 2).
Beta-blockers were required in 18% of women in our prospective cohort (Fig. 3). Discontinuation of ivabradine or switching to beta-blockers after recognition of pregnancy was well tolerated in most of our patients. Thus, necessity of ivabradine treatment should be critically considered in women with supraventricular tachycardia planning pregnancy. Data from previous studies suggest that discontinuation of ivabradine is possible in many (non-pregnant) patients with IST without recurrence. Even beneficial long-term effects after cessation of therapy are discussed [23, 24].
A high number of women in our cohort received therapy for anxiety and/or other psychiatric conditions. This is in line with findings from other studies reporting high rates of anxiety (25%) and depression (26%) in IST patients [25]. Anxiety is considered as non-cardiac symptom of inappropriate sinus tachycardia, and vice versa, anxiety may trigger tachycardia. Especially during pregnancy, empathic care and reassurance are necessary, to prevent increased concerns about possible harm to the fetus [26].
Preconception counselling is of utmost importance in all women considering pregnancy under medication [27]. Sinus tachycardia is a condition diagnosed mainly in young females and the first diagnosis is frequently made during or after pregnancy [25]. Considering the increasing off-label use of ivabradine as well as the high rate of > 40% unplanned pregnancies [28], more exposed pregnancies can be expected in the future. Well-established treatment options such as selected beta-blockers with low reproductive risk [29] should be considered in women of childbearing age before prescription of ivabradine.
Strength and limitations
A major strength of our study is the detailed exposure protocol and the multi-source ascertainment of pregnancy outcome via patients and their HCP followed by a case by case plausibility test.
Our data may not be representative of the German pregnant population; insofar, women with higher education are over-represented [30]. However, the regional distribution of enrolled patients is representative of the female population of childbearing age in Germany [11]. In addition, women on long-term ivabradine with severe cardiovascular conditions such as the approved treatment indications coronary artery disease and cardiac failure are under-represented in our cohort possibly favouring uneventful pregnancy outcomes. However, given the rarity of severe cardiac conditions in pregnant women, large study cohorts are not to be expected in the foreseeable future.
Conclusions
Our findings do not indicate that ivabradine is a major teratogen when used in early pregnancy. Although being the largest study published so far on the outcome of ivabradine-exposed pregnancies, the sample is too small to rule out embryotoxic effects. Further studies are needed to confirm or refute our findings.
In women planning pregnancy, established drugs of choice with low risk for the unborn should be preferred to ivabradine. For women with significant arrhythmias and severe cardiac disease, the impact of the underlying medical condition including comorbidities such as diabetes and high BMI on pregnancy outcome has to be considered when planning pregnancy.
After inadvertent ivabradine exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, all pregnant women with complicated arrhythmias or major cardiac disease should be carefully monitored.
Acknowledgements
We would like to thank all patients and physicians reporting pregnancies to the institute and acknowledge our colleagues from the German Embryotox Institute for counselling and thoroughly documenting cases, especially Sandra Zinke for continuous support with case documentation.
Contributions of authors
MH, ML, and CS designed the study. ML, EB, MH, and KD validated and analysed the data. ML, MH, and CS wrote the first draft of the manuscript, and all the authors critically revised subsequent manuscript drafts and contributed essential discussion points. All the authors approved the final version of this manuscript and they are responsible for the accuracy of this work.
Funding
Open Access funding enabled and organized by Projekt DEAL. This work was performed with financial support from the German Ministry of Health (BMG) and the German Federal Institute for Drugs and Medical Devices (BfArM). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
The study was registered with the German Clinical Trial register (DRKS00010502) and the study protocol was approved by the ethics committee of the Charité-Universitätsmedizin Berlin (EA1/107/16).
The study was registered with the German Clinical Trial register (DRKS00010502, 20.5.2016)
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | UNTIL GW 7+5 | DrugDosageText | CC BY | 33501507 | 19,157,674 | 2021-07 |
What was the outcome of reaction 'Atrial septal defect'? | Ivabradine use in pregnant women-treatment indications and pregnancy outcome: an evaluation of the German Embryotox database.
OBJECTIVE
Ivabradine has been approved for the treatment of chronic heart failure and chronic stable angina pectoris in Europe. Based on adverse outcomes of reproductive animal studies and the lack of human data, ivabradine is considered contraindicated during pregnancy. The aim of this observational study is to analyse ivabradine use before and during pregnancy.
METHODS
We evaluated all ivabradine-related requests to the German Embryotox Institute from 2007 to 2019. Exposed pregnancies were analysed as to their outcome.
RESULTS
Off-label use for supraventricular tachycardia was frequent in women of childbearing age. Of 38 prospectively ascertained pregnancies with ivabradine exposure and completed follow-up, 32 resulted in live births, 3 in spontaneous abortions, and 3 were electively terminated. One neonate presented with major birth defects (atrial septal defect and cleft palate). In 33/38 patients, ivabradine was discontinued after confirmation of pregnancy without cardiac deterioration and 5/38 women continued ivabradine throughout pregnancy. In addition, there were 3 retrospectively reported pregnancies including one major birth defect (tracheal atresia).
CONCLUSIONS
This case series represents the largest cohort of ivabradine-exposed pregnancies, published so far. According to our findings, ivabradine appears not to be a major teratogen. However, established drugs of choice with strong evidence of low risk for the unborn should be preferred in women planning pregnancy. After inadvertent exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, close monitoring is necessary in pregnant women with supraventricular arrhythmias or cardiac disease.
Introduction
Ivabradine is a heart rate-lowering agent, acting through selective inhibition of the If (I-funny) channel without additional hemodynamic effects on cardiac function. The approved treatment indications in Europe are chronic heart failure and symptomatic treatment of chronic stable angina pectoris after failing or intolerance of beta-blockers [1].
Preclinical studies have suggested fetotoxicity and teratogenic effects. In pregnant rats, ivabradine exposure equivalent to therapeutic levels in humans was associated with increased intrauterine and postnatal mortality. Ventricular septal defects and complex anomalies of the great arteries were noted at doses 3 times the therapeutic human exposure. In pregnant rabbits, reduced fetal and placental weight as well as ectrodactyly was observed at doses 15 times the therapeutic human exposure [2]. In addition, dose-dependent mortality was seen in chicken and mice embryos [3, 4].
ESC guidelines consider ivabradine contraindicated during pregnancy and recommend discontinuation independent of treatment indication before conception with close clinical and echocardiographic monitoring [5].
Published data on ivabradine-exposed pregnancies are limited to three case reports, one with ivabradine treatment in the late first and early second trimester for sinus tachycardia after myocardial infarction [6], and the second starting ivabradine in the second trimester for deteriorating tachycardia-induced cardiomyopathy [7]. In the third case with first trimester exposure, pregnancy was electively terminated [8].
The limited human experience along with the teratogenic effects in animal studies makes it difficult to counsel inadvertently exposed pregnant women and their health care providers (HCP). Therefore, we decided to evaluate ivabradine-exposed pregnancies recorded by the German Embryotox Institute.
Methods
Study design
The German Embryotox Institute offers risk assessment on drug exposure during pregnancy to HCPs and patients. Up to 15,000 annual requests are answered, and in approximately 3500 critically exposed cases per year, the course and outcome of pregnancy are documented. At the initial contact, maternal characteristics and detailed exposure assessment are asked for after informed consent. About 8 weeks after the estimated date of birth, information on course and outcome of pregnancy is collected via structured telephone interview and mailed questionnaires. After a case by case plausibility check, additional information and health care records are requested in cases of adverse outcomes or inconsistent information.
Weeks of pregnancy were based on ultrasound and if not available on the last menstrual period (LMP). First trimester was defined as gestational week 2 + 0 (conception) until 12 + 6. Major birth defects were classified according to EUROCAT [9]. In prospectively ascertained pregnancies, neither the outcome of pregnancy nor prenatal pathology was known at the time of first contact. Pregnancies that were reported after birth or because of prenatal pathology were considered retrospective and evaluated separately. A detailed description of methodology adapted to the recommendations of the Strengthening of the Reporting of Observational studies in Epidemiology (STROBE) statement is given in Schaefer et al. [10] and Dathe and Schaefer [11].
We evaluated all requests on ivabradine to the German Embryotox Institute from 2007 to 2019.
Statistical analyses
Descriptive statistics were applied. For age, BMI, gestational weeks, neonatal weight, length, and head circumference, median and interquartile ranges are presented. Birth weights were adjusted to sex and gestational age at birth and percentile categories and standard deviation scores (SDS) were calculated according to the German perinatal survey [12].
Results
During the study period 2007–2019, our institute received 97 requests for ivabradine, 56 (58%) were from HCP, and 41 from patients. For further details, see Fig. 1. Treatment indications for ivabradine over time in relation to formal approval status and pregnancy labelling are summarized in Fig. 2. Supraventricular tachycardia was the most common reason for ivabradine treatment in our cohort. The first ivabradine-exposed pregnancy was recorded in March 2007, 1 year after approval and marketing authorization in Germany in January 2006.Fig. 1 Overview on information requests to the German Embryotox institute on ivabradine
Fig. 2 Treatment indication for ivabradine in 97 Embryotox requests over time in relation to pregnancy labelling and approval status by EMA/FDA. Recommendations for use of ivabradine in supraventricular tachycardias, i.e. inappropriate sinus tachycardia and postural orthostatic tachycardia syndrome: HRS/ESC 2015 [13], ACC/AHA/HFSA 2016 [14], ESC 2019 [15]. ACC, American College of Cardiology; EMA, European Medicines Agency; ESC, European Society of Cardiology; FDA, US Food and Drug Administration; HRS, Heart Rhythm Society
Fifty-six of the 97 requests were related to maternal exposure during pregnancy (Fig. 1). In 41 of these cases, follow-up on pregnancy outcome could be completed: 38 pregnancies were prospectively ascertained and three were retrospective reports.
Evaluation of prospectively ascertained pregnancies
Maternal characteristics
The median age of pregnant women was 29.5 years and the median BMI 22.5. Maternal characteristics are presented in detail in Table 1. The mean time at first contact was gestational week (GW) 7. Only 16 of the 38 prospectively ascertained pregnancies were actually planned (42%) and only 7 women (18%) started folic acid before conception. Contraceptive methods, mostly oral contraceptives (n = 4), were still used by 6 women at conception.Table 1 Maternal characteristics and pregnancy outcomes of ivabradine exposed pregnancies. Data are n (%) except for age, BMI, gestational week at birth, neonatal weight, length and head circumference, which are median and interquartile range
Maternal characteristics N (%) or median (IQR)
Maternal characteristics Age (n=38) 29.5 (26.5-34)
BMI (n=34) 22.5 (19.4-25.6)
Smoking (n=37)
>5 cigarettes/day 7 (19%)
<=5 cigarettes/day 4 (11%)
Alcohol (n=36)
<=1 drink/day 2 (6%)
>1 drink/day 1 (3%)
GW at first contact (n=38) 7 (5.6-11.1)
Previous pregnancies (n=38)
0 17(45%)
1 8 (21%)
2 13 (34%)
Pregnancy outcome and neonatal characteristics N (%) or median (IQR)
Pregnancy outcome (n=38) Spontaneous abortion 3
ETOP 3
Live birth 32
Pregnancy complications (n=32, only live births) Gestational diabetes 5 (16%)
Pre-eclampsia 2 (6%)
Cesarean section 15 (46%)
Preterm birth (<GW 37) 3 (9%)
Neonatal characteristics (n=33, including one pair of twins) GW at birth (n=32) 39 (37.9-39.9)
Weight, g (n=32) 3300 (2935-3610)
Length, cm (n=31) 50 (49-52.5)
HC, cm (n=30) 35 (34-36)
SGA (n=32) 2 (6%)
BMI, Body Mass Index; ETOP, elective termination of pregnancy; GW, gestational week; HC, head circumference; IQR, interquartile range; SGA, small for gestational age (<10th percentile)
Ivabradine exposure and co-medication
The most prevalent indication for ivabradine treatment was supraventricular tachycardia (n = 29) followed by heart failure (n = 3) and coronary artery disease (n = 2). Other reported indications (n = 4) included QT prolongation and sick-sinus-syndrome, although both are regarded as contraindications for ivabradine. The median daily ivabradine dose was 7.5 mg (IQR 5–10, min–max 2.5–15, n = 33).
Nearly all women were treated with ivabradine during the first trimester (n = 37) and 32 stopped treatment after recognition of pregnancy (Fig. 3), at median gestational week 6.1 (interquartile range 5.4–7.8). Cardiac decompensation or complications were not described after cessation of treatment. Only 5 women continued ivabradine throughout pregnancy or until shortly before delivery. One woman started treatment after first trimester (GW 18) of her twin pregnancy due to symptomatic tachycardia under beta-blocker therapy (Iv34 in Fig. 3).Fig. 3 Ivabradine exposure and pregnancy outcome of prospectively (n = 38) and retrospectively (n = 3) ascertained pregnancies
Cardiovascular co-medication consisted of beta-blockers (mainly metoprolol) in 15 women, mostly concomitantly (n = 7) or in replacement of ivabradine (n = 6). Two women initiated ivabradine after discontinuation of beta-blockers (Iv30 and Iv34 in Fig. 3). Six women received other cardiovascular medication, three methyldopa or amlodipine for hypertension, two digitoxin or verapamil for tachyarrhythmia, and one sildenafil and eplerenone for cardiomyopathy with cardiac failure (Iv17).
Psychiatric co-medication was reported in 7/38 women (5 of these with an anxiety disorder) including the teratogenic antiepileptic topiramate (n = 1) and pregabalin (n = 3).
Pregnancy outcome
Of the 38 prospectively ascertained ivabradine-exposed pregnancies, 32 were live births (33 infants including one pair of twins) and three resulted in early spontaneous abortions (Iv10, Iv17, and Iv19 in Fig. 3). One of these occurred in a woman with cardiomyopathy (Iv17) and various co-medications including hydroxycarbamide for thrombocythemia until GW 7. Three pregnancies were electively terminated for social/psychological reasons. Further information on pregnancy complications and outcomes is given in Table 1 and Fig. 3.
Major birth defects were reported in one pregnancy (Iv24) with 15 mg ivabradine plus bisoprolol until GW 7 + 5, both replaced by metoprolol. Polyhydramnios and preeclampsia led to Caesarean section in GW 35 after premature rupture of membranes. The premature child was large for gestational age (> 97 percentile) and had a cleft palate, an atrial septal defect II (spontaneously closed at the age of 10 months), and a hypertrophic cardiac septum. The mother was treated for sinus tachycardia, in addition she had diabetes type I and a BMI of 30.
Neonatal characteristics
Characteristics of the 33 neonates are summarized in Table 1. Compared to the German perinatal survey, standardized birth weights were lower with a median SDS of − 0.36, corresponding to a reduction of birth weight of 149 g for singletons born in gestational week 40.
Two neonates were small for gestational age, one of the twins (exposure only during 2nd and 3rd trimester, Iv34) and one after maternal ivabradine exposure until GW 12 (Iv29). Two neonates were large for gestational age, both mothers were diabetic (Iv24, type I diabetes and Iv27, gestational diabetes).
Retrospective cases
During the study period three pregnancies were retrospectively reported (further details are given in Table 2 and Fig. 3). One premature neonate (IvR1) with tracheal atresia died perinatally. The mother had been treated with ivabradine, naproxen, and metamizole during the first trimester until recognition of pregnancy (no further details available).Table 2 Retrospectively reported pregnancies
Case Maternal age/BMI Treatment indication Ivabradine exposure, other co-medication (GW or trimester) Pregnancy outcome, GW, birth weight percentile, sex Pregnancy complications, birth defects, and additional findings
IvR1 31 years/31.2 Tachycardia Ivabradine (1st trim)
Naproxen (1st trim)
Metamizole (1st trim)
Live birth, 36 + 2, > 97th, f Polyhydramnios, perinatal death, tracheal atresia, normal karyotype (46, XX)
IvR2 29 years/18.4 Tachycardia, syncopes, pacemaker Ivabradine (1st trim)
Bisoprolol (0–35 + 5)
Levetiracetam (0–35 + 5, grand mal epilepsy)
Live birth, 35 + 5, < 10th, f IUGR (since GW 30), oligohydramnios (since GW 34), Caesarean section (vaginal bleeding)
IvR3 29 years/not available Tachycardia Ivabradine (0–41 + 3)
Bisoprolol (0–41 + 3)
Live birth, 41 + 3
> 10th, m
Uncomplicated pregnancy
BMI, body mass index; f, female; GW, gestational week; IUGR, intrauterine growth restriction; m, male; trim, trimester
The other two retrospective cases include a woman with comorbidities, intrauterine growth restriction, and premature birth (IvR2). The second was a healthy newborn (IvR3) after treatment with ivabradine throughout pregnancy. This mother reported a previous also uneventful pregnancy course under ivabradine.
Discussion
Pregnancy outcome
In our prospective cohort, there was only one neonate with major birth defects (cleft palate and atrial septal defect) from a diabetic mother with supraventricular tachycardia. No further major or cardiac malformations were reported among the 32 live births and 6 pregnancy losses. Cardiac defects have been observed in preclinical experimental studies. As a possible teratogenic mechanism, ivabradine-induced embryonic heart rate reduction leading to reduced cardiac output and hypoxia was discussed [16]. With a prevalence of almost 1 of 100 cardiac malformations belong to the most frequent birth defects [17, 18]. In our case, the neonate was large for gestational age and had a septum hypertrophy, supporting the assumption that maternal pre-gestational diabetes may also be causative for the observed cardiac defect and cleft palate. A causal relationship between ivabradine and cleft palate is unlikely, because ivabradine was discontinued at GW 7 + 5 before the beginning of the critical time period for the development of cleft palate at gestational week 9. The rate of major birth defects in our prospective cohort is comparable to the expected background rate of 2–3% [17, 18]. However, the limited number of pregnancies exposed during the entire period of embryogenesis does not allow excluding an increased risk, so far.
There was only one retrospectively reported case with a birth defect (malformation of the trachea) suspected as potentially ivabradine related. The concomitant exposures with naproxen and metamizole have not been evidenced as teratogens in human, whereas the high maternal BMI in the present case poses a risk for birth defects.
Taken together, we did not observe a specific pattern of congenital anomalies in our study. In addition, the presence of maternal comorbidities like diabetes and high BMI may have contributed to the observed birth defects.
The number of spontaneous abortions (3/38) was not higher than expected. The rate of preeclampsia and preterm birth in our cohort was within the expected range [19].
Treatment indications
Although the approved treatment indications for ivabradine, chronic heart failure and chronic stable angina pectoris (see Fig. 2 for changes of licenced use during the study period), are rare in women of reproductive age, we have observed an increasing need for information for off-label use in (younger) women considering pregnancy.
In our case series, most women were treated (off-label) for supraventricular tachycardia using lower doses than recommended (5–7.5 mg twice daily). Supraventricular tachycardia is not uncommon, especially in younger women, and is expected in 0.02–0.5% of pregnancies [20]. Aggravation of symptoms during pregnancy is described in about 20–50% of women related to various mechanisms, including pregnancy-related increase in cardiac heart rate and output and hormonal changes [20].
In 2015, the HRS (Heart Rhythm Society) considered ivabradine as promising for treating patients with inappropriate sinus tachycardia (IST) [13], followed by a recommendation for ivabradine by the ACC (American College of Cardiology) in 2016 [14]—though only one small randomized controlled study has yet supported beneficial effects [21]. Although still not approved for this indication, recent ESC guidelines [15] also recommend ivabradine alone or in combination with a beta-blocker in symptomatic patients with IST, for the treatment of focal atrial tachycardia or for postural orthostatic tachycardia syndrome (POTS), a condition considered as common cause for orthostatic intolerance in women between 15 and 25 years [20, 22] (Fig. 2).
Beta-blockers were required in 18% of women in our prospective cohort (Fig. 3). Discontinuation of ivabradine or switching to beta-blockers after recognition of pregnancy was well tolerated in most of our patients. Thus, necessity of ivabradine treatment should be critically considered in women with supraventricular tachycardia planning pregnancy. Data from previous studies suggest that discontinuation of ivabradine is possible in many (non-pregnant) patients with IST without recurrence. Even beneficial long-term effects after cessation of therapy are discussed [23, 24].
A high number of women in our cohort received therapy for anxiety and/or other psychiatric conditions. This is in line with findings from other studies reporting high rates of anxiety (25%) and depression (26%) in IST patients [25]. Anxiety is considered as non-cardiac symptom of inappropriate sinus tachycardia, and vice versa, anxiety may trigger tachycardia. Especially during pregnancy, empathic care and reassurance are necessary, to prevent increased concerns about possible harm to the fetus [26].
Preconception counselling is of utmost importance in all women considering pregnancy under medication [27]. Sinus tachycardia is a condition diagnosed mainly in young females and the first diagnosis is frequently made during or after pregnancy [25]. Considering the increasing off-label use of ivabradine as well as the high rate of > 40% unplanned pregnancies [28], more exposed pregnancies can be expected in the future. Well-established treatment options such as selected beta-blockers with low reproductive risk [29] should be considered in women of childbearing age before prescription of ivabradine.
Strength and limitations
A major strength of our study is the detailed exposure protocol and the multi-source ascertainment of pregnancy outcome via patients and their HCP followed by a case by case plausibility test.
Our data may not be representative of the German pregnant population; insofar, women with higher education are over-represented [30]. However, the regional distribution of enrolled patients is representative of the female population of childbearing age in Germany [11]. In addition, women on long-term ivabradine with severe cardiovascular conditions such as the approved treatment indications coronary artery disease and cardiac failure are under-represented in our cohort possibly favouring uneventful pregnancy outcomes. However, given the rarity of severe cardiac conditions in pregnant women, large study cohorts are not to be expected in the foreseeable future.
Conclusions
Our findings do not indicate that ivabradine is a major teratogen when used in early pregnancy. Although being the largest study published so far on the outcome of ivabradine-exposed pregnancies, the sample is too small to rule out embryotoxic effects. Further studies are needed to confirm or refute our findings.
In women planning pregnancy, established drugs of choice with low risk for the unborn should be preferred to ivabradine. For women with significant arrhythmias and severe cardiac disease, the impact of the underlying medical condition including comorbidities such as diabetes and high BMI on pregnancy outcome has to be considered when planning pregnancy.
After inadvertent ivabradine exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, all pregnant women with complicated arrhythmias or major cardiac disease should be carefully monitored.
Acknowledgements
We would like to thank all patients and physicians reporting pregnancies to the institute and acknowledge our colleagues from the German Embryotox Institute for counselling and thoroughly documenting cases, especially Sandra Zinke for continuous support with case documentation.
Contributions of authors
MH, ML, and CS designed the study. ML, EB, MH, and KD validated and analysed the data. ML, MH, and CS wrote the first draft of the manuscript, and all the authors critically revised subsequent manuscript drafts and contributed essential discussion points. All the authors approved the final version of this manuscript and they are responsible for the accuracy of this work.
Funding
Open Access funding enabled and organized by Projekt DEAL. This work was performed with financial support from the German Ministry of Health (BMG) and the German Federal Institute for Drugs and Medical Devices (BfArM). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
The study was registered with the German Clinical Trial register (DRKS00010502) and the study protocol was approved by the ethics committee of the Charité-Universitätsmedizin Berlin (EA1/107/16).
The study was registered with the German Clinical Trial register (DRKS00010502, 20.5.2016)
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Recovered | ReactionOutcome | CC BY | 33501507 | 19,697,614 | 2021-07 |
What was the outcome of reaction 'Exposure during pregnancy'? | Ivabradine use in pregnant women-treatment indications and pregnancy outcome: an evaluation of the German Embryotox database.
OBJECTIVE
Ivabradine has been approved for the treatment of chronic heart failure and chronic stable angina pectoris in Europe. Based on adverse outcomes of reproductive animal studies and the lack of human data, ivabradine is considered contraindicated during pregnancy. The aim of this observational study is to analyse ivabradine use before and during pregnancy.
METHODS
We evaluated all ivabradine-related requests to the German Embryotox Institute from 2007 to 2019. Exposed pregnancies were analysed as to their outcome.
RESULTS
Off-label use for supraventricular tachycardia was frequent in women of childbearing age. Of 38 prospectively ascertained pregnancies with ivabradine exposure and completed follow-up, 32 resulted in live births, 3 in spontaneous abortions, and 3 were electively terminated. One neonate presented with major birth defects (atrial septal defect and cleft palate). In 33/38 patients, ivabradine was discontinued after confirmation of pregnancy without cardiac deterioration and 5/38 women continued ivabradine throughout pregnancy. In addition, there were 3 retrospectively reported pregnancies including one major birth defect (tracheal atresia).
CONCLUSIONS
This case series represents the largest cohort of ivabradine-exposed pregnancies, published so far. According to our findings, ivabradine appears not to be a major teratogen. However, established drugs of choice with strong evidence of low risk for the unborn should be preferred in women planning pregnancy. After inadvertent exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, close monitoring is necessary in pregnant women with supraventricular arrhythmias or cardiac disease.
Introduction
Ivabradine is a heart rate-lowering agent, acting through selective inhibition of the If (I-funny) channel without additional hemodynamic effects on cardiac function. The approved treatment indications in Europe are chronic heart failure and symptomatic treatment of chronic stable angina pectoris after failing or intolerance of beta-blockers [1].
Preclinical studies have suggested fetotoxicity and teratogenic effects. In pregnant rats, ivabradine exposure equivalent to therapeutic levels in humans was associated with increased intrauterine and postnatal mortality. Ventricular septal defects and complex anomalies of the great arteries were noted at doses 3 times the therapeutic human exposure. In pregnant rabbits, reduced fetal and placental weight as well as ectrodactyly was observed at doses 15 times the therapeutic human exposure [2]. In addition, dose-dependent mortality was seen in chicken and mice embryos [3, 4].
ESC guidelines consider ivabradine contraindicated during pregnancy and recommend discontinuation independent of treatment indication before conception with close clinical and echocardiographic monitoring [5].
Published data on ivabradine-exposed pregnancies are limited to three case reports, one with ivabradine treatment in the late first and early second trimester for sinus tachycardia after myocardial infarction [6], and the second starting ivabradine in the second trimester for deteriorating tachycardia-induced cardiomyopathy [7]. In the third case with first trimester exposure, pregnancy was electively terminated [8].
The limited human experience along with the teratogenic effects in animal studies makes it difficult to counsel inadvertently exposed pregnant women and their health care providers (HCP). Therefore, we decided to evaluate ivabradine-exposed pregnancies recorded by the German Embryotox Institute.
Methods
Study design
The German Embryotox Institute offers risk assessment on drug exposure during pregnancy to HCPs and patients. Up to 15,000 annual requests are answered, and in approximately 3500 critically exposed cases per year, the course and outcome of pregnancy are documented. At the initial contact, maternal characteristics and detailed exposure assessment are asked for after informed consent. About 8 weeks after the estimated date of birth, information on course and outcome of pregnancy is collected via structured telephone interview and mailed questionnaires. After a case by case plausibility check, additional information and health care records are requested in cases of adverse outcomes or inconsistent information.
Weeks of pregnancy were based on ultrasound and if not available on the last menstrual period (LMP). First trimester was defined as gestational week 2 + 0 (conception) until 12 + 6. Major birth defects were classified according to EUROCAT [9]. In prospectively ascertained pregnancies, neither the outcome of pregnancy nor prenatal pathology was known at the time of first contact. Pregnancies that were reported after birth or because of prenatal pathology were considered retrospective and evaluated separately. A detailed description of methodology adapted to the recommendations of the Strengthening of the Reporting of Observational studies in Epidemiology (STROBE) statement is given in Schaefer et al. [10] and Dathe and Schaefer [11].
We evaluated all requests on ivabradine to the German Embryotox Institute from 2007 to 2019.
Statistical analyses
Descriptive statistics were applied. For age, BMI, gestational weeks, neonatal weight, length, and head circumference, median and interquartile ranges are presented. Birth weights were adjusted to sex and gestational age at birth and percentile categories and standard deviation scores (SDS) were calculated according to the German perinatal survey [12].
Results
During the study period 2007–2019, our institute received 97 requests for ivabradine, 56 (58%) were from HCP, and 41 from patients. For further details, see Fig. 1. Treatment indications for ivabradine over time in relation to formal approval status and pregnancy labelling are summarized in Fig. 2. Supraventricular tachycardia was the most common reason for ivabradine treatment in our cohort. The first ivabradine-exposed pregnancy was recorded in March 2007, 1 year after approval and marketing authorization in Germany in January 2006.Fig. 1 Overview on information requests to the German Embryotox institute on ivabradine
Fig. 2 Treatment indication for ivabradine in 97 Embryotox requests over time in relation to pregnancy labelling and approval status by EMA/FDA. Recommendations for use of ivabradine in supraventricular tachycardias, i.e. inappropriate sinus tachycardia and postural orthostatic tachycardia syndrome: HRS/ESC 2015 [13], ACC/AHA/HFSA 2016 [14], ESC 2019 [15]. ACC, American College of Cardiology; EMA, European Medicines Agency; ESC, European Society of Cardiology; FDA, US Food and Drug Administration; HRS, Heart Rhythm Society
Fifty-six of the 97 requests were related to maternal exposure during pregnancy (Fig. 1). In 41 of these cases, follow-up on pregnancy outcome could be completed: 38 pregnancies were prospectively ascertained and three were retrospective reports.
Evaluation of prospectively ascertained pregnancies
Maternal characteristics
The median age of pregnant women was 29.5 years and the median BMI 22.5. Maternal characteristics are presented in detail in Table 1. The mean time at first contact was gestational week (GW) 7. Only 16 of the 38 prospectively ascertained pregnancies were actually planned (42%) and only 7 women (18%) started folic acid before conception. Contraceptive methods, mostly oral contraceptives (n = 4), were still used by 6 women at conception.Table 1 Maternal characteristics and pregnancy outcomes of ivabradine exposed pregnancies. Data are n (%) except for age, BMI, gestational week at birth, neonatal weight, length and head circumference, which are median and interquartile range
Maternal characteristics N (%) or median (IQR)
Maternal characteristics Age (n=38) 29.5 (26.5-34)
BMI (n=34) 22.5 (19.4-25.6)
Smoking (n=37)
>5 cigarettes/day 7 (19%)
<=5 cigarettes/day 4 (11%)
Alcohol (n=36)
<=1 drink/day 2 (6%)
>1 drink/day 1 (3%)
GW at first contact (n=38) 7 (5.6-11.1)
Previous pregnancies (n=38)
0 17(45%)
1 8 (21%)
2 13 (34%)
Pregnancy outcome and neonatal characteristics N (%) or median (IQR)
Pregnancy outcome (n=38) Spontaneous abortion 3
ETOP 3
Live birth 32
Pregnancy complications (n=32, only live births) Gestational diabetes 5 (16%)
Pre-eclampsia 2 (6%)
Cesarean section 15 (46%)
Preterm birth (<GW 37) 3 (9%)
Neonatal characteristics (n=33, including one pair of twins) GW at birth (n=32) 39 (37.9-39.9)
Weight, g (n=32) 3300 (2935-3610)
Length, cm (n=31) 50 (49-52.5)
HC, cm (n=30) 35 (34-36)
SGA (n=32) 2 (6%)
BMI, Body Mass Index; ETOP, elective termination of pregnancy; GW, gestational week; HC, head circumference; IQR, interquartile range; SGA, small for gestational age (<10th percentile)
Ivabradine exposure and co-medication
The most prevalent indication for ivabradine treatment was supraventricular tachycardia (n = 29) followed by heart failure (n = 3) and coronary artery disease (n = 2). Other reported indications (n = 4) included QT prolongation and sick-sinus-syndrome, although both are regarded as contraindications for ivabradine. The median daily ivabradine dose was 7.5 mg (IQR 5–10, min–max 2.5–15, n = 33).
Nearly all women were treated with ivabradine during the first trimester (n = 37) and 32 stopped treatment after recognition of pregnancy (Fig. 3), at median gestational week 6.1 (interquartile range 5.4–7.8). Cardiac decompensation or complications were not described after cessation of treatment. Only 5 women continued ivabradine throughout pregnancy or until shortly before delivery. One woman started treatment after first trimester (GW 18) of her twin pregnancy due to symptomatic tachycardia under beta-blocker therapy (Iv34 in Fig. 3).Fig. 3 Ivabradine exposure and pregnancy outcome of prospectively (n = 38) and retrospectively (n = 3) ascertained pregnancies
Cardiovascular co-medication consisted of beta-blockers (mainly metoprolol) in 15 women, mostly concomitantly (n = 7) or in replacement of ivabradine (n = 6). Two women initiated ivabradine after discontinuation of beta-blockers (Iv30 and Iv34 in Fig. 3). Six women received other cardiovascular medication, three methyldopa or amlodipine for hypertension, two digitoxin or verapamil for tachyarrhythmia, and one sildenafil and eplerenone for cardiomyopathy with cardiac failure (Iv17).
Psychiatric co-medication was reported in 7/38 women (5 of these with an anxiety disorder) including the teratogenic antiepileptic topiramate (n = 1) and pregabalin (n = 3).
Pregnancy outcome
Of the 38 prospectively ascertained ivabradine-exposed pregnancies, 32 were live births (33 infants including one pair of twins) and three resulted in early spontaneous abortions (Iv10, Iv17, and Iv19 in Fig. 3). One of these occurred in a woman with cardiomyopathy (Iv17) and various co-medications including hydroxycarbamide for thrombocythemia until GW 7. Three pregnancies were electively terminated for social/psychological reasons. Further information on pregnancy complications and outcomes is given in Table 1 and Fig. 3.
Major birth defects were reported in one pregnancy (Iv24) with 15 mg ivabradine plus bisoprolol until GW 7 + 5, both replaced by metoprolol. Polyhydramnios and preeclampsia led to Caesarean section in GW 35 after premature rupture of membranes. The premature child was large for gestational age (> 97 percentile) and had a cleft palate, an atrial septal defect II (spontaneously closed at the age of 10 months), and a hypertrophic cardiac septum. The mother was treated for sinus tachycardia, in addition she had diabetes type I and a BMI of 30.
Neonatal characteristics
Characteristics of the 33 neonates are summarized in Table 1. Compared to the German perinatal survey, standardized birth weights were lower with a median SDS of − 0.36, corresponding to a reduction of birth weight of 149 g for singletons born in gestational week 40.
Two neonates were small for gestational age, one of the twins (exposure only during 2nd and 3rd trimester, Iv34) and one after maternal ivabradine exposure until GW 12 (Iv29). Two neonates were large for gestational age, both mothers were diabetic (Iv24, type I diabetes and Iv27, gestational diabetes).
Retrospective cases
During the study period three pregnancies were retrospectively reported (further details are given in Table 2 and Fig. 3). One premature neonate (IvR1) with tracheal atresia died perinatally. The mother had been treated with ivabradine, naproxen, and metamizole during the first trimester until recognition of pregnancy (no further details available).Table 2 Retrospectively reported pregnancies
Case Maternal age/BMI Treatment indication Ivabradine exposure, other co-medication (GW or trimester) Pregnancy outcome, GW, birth weight percentile, sex Pregnancy complications, birth defects, and additional findings
IvR1 31 years/31.2 Tachycardia Ivabradine (1st trim)
Naproxen (1st trim)
Metamizole (1st trim)
Live birth, 36 + 2, > 97th, f Polyhydramnios, perinatal death, tracheal atresia, normal karyotype (46, XX)
IvR2 29 years/18.4 Tachycardia, syncopes, pacemaker Ivabradine (1st trim)
Bisoprolol (0–35 + 5)
Levetiracetam (0–35 + 5, grand mal epilepsy)
Live birth, 35 + 5, < 10th, f IUGR (since GW 30), oligohydramnios (since GW 34), Caesarean section (vaginal bleeding)
IvR3 29 years/not available Tachycardia Ivabradine (0–41 + 3)
Bisoprolol (0–41 + 3)
Live birth, 41 + 3
> 10th, m
Uncomplicated pregnancy
BMI, body mass index; f, female; GW, gestational week; IUGR, intrauterine growth restriction; m, male; trim, trimester
The other two retrospective cases include a woman with comorbidities, intrauterine growth restriction, and premature birth (IvR2). The second was a healthy newborn (IvR3) after treatment with ivabradine throughout pregnancy. This mother reported a previous also uneventful pregnancy course under ivabradine.
Discussion
Pregnancy outcome
In our prospective cohort, there was only one neonate with major birth defects (cleft palate and atrial septal defect) from a diabetic mother with supraventricular tachycardia. No further major or cardiac malformations were reported among the 32 live births and 6 pregnancy losses. Cardiac defects have been observed in preclinical experimental studies. As a possible teratogenic mechanism, ivabradine-induced embryonic heart rate reduction leading to reduced cardiac output and hypoxia was discussed [16]. With a prevalence of almost 1 of 100 cardiac malformations belong to the most frequent birth defects [17, 18]. In our case, the neonate was large for gestational age and had a septum hypertrophy, supporting the assumption that maternal pre-gestational diabetes may also be causative for the observed cardiac defect and cleft palate. A causal relationship between ivabradine and cleft palate is unlikely, because ivabradine was discontinued at GW 7 + 5 before the beginning of the critical time period for the development of cleft palate at gestational week 9. The rate of major birth defects in our prospective cohort is comparable to the expected background rate of 2–3% [17, 18]. However, the limited number of pregnancies exposed during the entire period of embryogenesis does not allow excluding an increased risk, so far.
There was only one retrospectively reported case with a birth defect (malformation of the trachea) suspected as potentially ivabradine related. The concomitant exposures with naproxen and metamizole have not been evidenced as teratogens in human, whereas the high maternal BMI in the present case poses a risk for birth defects.
Taken together, we did not observe a specific pattern of congenital anomalies in our study. In addition, the presence of maternal comorbidities like diabetes and high BMI may have contributed to the observed birth defects.
The number of spontaneous abortions (3/38) was not higher than expected. The rate of preeclampsia and preterm birth in our cohort was within the expected range [19].
Treatment indications
Although the approved treatment indications for ivabradine, chronic heart failure and chronic stable angina pectoris (see Fig. 2 for changes of licenced use during the study period), are rare in women of reproductive age, we have observed an increasing need for information for off-label use in (younger) women considering pregnancy.
In our case series, most women were treated (off-label) for supraventricular tachycardia using lower doses than recommended (5–7.5 mg twice daily). Supraventricular tachycardia is not uncommon, especially in younger women, and is expected in 0.02–0.5% of pregnancies [20]. Aggravation of symptoms during pregnancy is described in about 20–50% of women related to various mechanisms, including pregnancy-related increase in cardiac heart rate and output and hormonal changes [20].
In 2015, the HRS (Heart Rhythm Society) considered ivabradine as promising for treating patients with inappropriate sinus tachycardia (IST) [13], followed by a recommendation for ivabradine by the ACC (American College of Cardiology) in 2016 [14]—though only one small randomized controlled study has yet supported beneficial effects [21]. Although still not approved for this indication, recent ESC guidelines [15] also recommend ivabradine alone or in combination with a beta-blocker in symptomatic patients with IST, for the treatment of focal atrial tachycardia or for postural orthostatic tachycardia syndrome (POTS), a condition considered as common cause for orthostatic intolerance in women between 15 and 25 years [20, 22] (Fig. 2).
Beta-blockers were required in 18% of women in our prospective cohort (Fig. 3). Discontinuation of ivabradine or switching to beta-blockers after recognition of pregnancy was well tolerated in most of our patients. Thus, necessity of ivabradine treatment should be critically considered in women with supraventricular tachycardia planning pregnancy. Data from previous studies suggest that discontinuation of ivabradine is possible in many (non-pregnant) patients with IST without recurrence. Even beneficial long-term effects after cessation of therapy are discussed [23, 24].
A high number of women in our cohort received therapy for anxiety and/or other psychiatric conditions. This is in line with findings from other studies reporting high rates of anxiety (25%) and depression (26%) in IST patients [25]. Anxiety is considered as non-cardiac symptom of inappropriate sinus tachycardia, and vice versa, anxiety may trigger tachycardia. Especially during pregnancy, empathic care and reassurance are necessary, to prevent increased concerns about possible harm to the fetus [26].
Preconception counselling is of utmost importance in all women considering pregnancy under medication [27]. Sinus tachycardia is a condition diagnosed mainly in young females and the first diagnosis is frequently made during or after pregnancy [25]. Considering the increasing off-label use of ivabradine as well as the high rate of > 40% unplanned pregnancies [28], more exposed pregnancies can be expected in the future. Well-established treatment options such as selected beta-blockers with low reproductive risk [29] should be considered in women of childbearing age before prescription of ivabradine.
Strength and limitations
A major strength of our study is the detailed exposure protocol and the multi-source ascertainment of pregnancy outcome via patients and their HCP followed by a case by case plausibility test.
Our data may not be representative of the German pregnant population; insofar, women with higher education are over-represented [30]. However, the regional distribution of enrolled patients is representative of the female population of childbearing age in Germany [11]. In addition, women on long-term ivabradine with severe cardiovascular conditions such as the approved treatment indications coronary artery disease and cardiac failure are under-represented in our cohort possibly favouring uneventful pregnancy outcomes. However, given the rarity of severe cardiac conditions in pregnant women, large study cohorts are not to be expected in the foreseeable future.
Conclusions
Our findings do not indicate that ivabradine is a major teratogen when used in early pregnancy. Although being the largest study published so far on the outcome of ivabradine-exposed pregnancies, the sample is too small to rule out embryotoxic effects. Further studies are needed to confirm or refute our findings.
In women planning pregnancy, established drugs of choice with low risk for the unborn should be preferred to ivabradine. For women with significant arrhythmias and severe cardiac disease, the impact of the underlying medical condition including comorbidities such as diabetes and high BMI on pregnancy outcome has to be considered when planning pregnancy.
After inadvertent ivabradine exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, all pregnant women with complicated arrhythmias or major cardiac disease should be carefully monitored.
Acknowledgements
We would like to thank all patients and physicians reporting pregnancies to the institute and acknowledge our colleagues from the German Embryotox Institute for counselling and thoroughly documenting cases, especially Sandra Zinke for continuous support with case documentation.
Contributions of authors
MH, ML, and CS designed the study. ML, EB, MH, and KD validated and analysed the data. ML, MH, and CS wrote the first draft of the manuscript, and all the authors critically revised subsequent manuscript drafts and contributed essential discussion points. All the authors approved the final version of this manuscript and they are responsible for the accuracy of this work.
Funding
Open Access funding enabled and organized by Projekt DEAL. This work was performed with financial support from the German Ministry of Health (BMG) and the German Federal Institute for Drugs and Medical Devices (BfArM). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
The study was registered with the German Clinical Trial register (DRKS00010502) and the study protocol was approved by the ethics committee of the Charité-Universitätsmedizin Berlin (EA1/107/16).
The study was registered with the German Clinical Trial register (DRKS00010502, 20.5.2016)
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Recovered | ReactionOutcome | CC BY | 33501507 | 19,157,674 | 2021-07 |
What was the outcome of reaction 'Maternal exposure during pregnancy'? | Ivabradine use in pregnant women-treatment indications and pregnancy outcome: an evaluation of the German Embryotox database.
OBJECTIVE
Ivabradine has been approved for the treatment of chronic heart failure and chronic stable angina pectoris in Europe. Based on adverse outcomes of reproductive animal studies and the lack of human data, ivabradine is considered contraindicated during pregnancy. The aim of this observational study is to analyse ivabradine use before and during pregnancy.
METHODS
We evaluated all ivabradine-related requests to the German Embryotox Institute from 2007 to 2019. Exposed pregnancies were analysed as to their outcome.
RESULTS
Off-label use for supraventricular tachycardia was frequent in women of childbearing age. Of 38 prospectively ascertained pregnancies with ivabradine exposure and completed follow-up, 32 resulted in live births, 3 in spontaneous abortions, and 3 were electively terminated. One neonate presented with major birth defects (atrial septal defect and cleft palate). In 33/38 patients, ivabradine was discontinued after confirmation of pregnancy without cardiac deterioration and 5/38 women continued ivabradine throughout pregnancy. In addition, there were 3 retrospectively reported pregnancies including one major birth defect (tracheal atresia).
CONCLUSIONS
This case series represents the largest cohort of ivabradine-exposed pregnancies, published so far. According to our findings, ivabradine appears not to be a major teratogen. However, established drugs of choice with strong evidence of low risk for the unborn should be preferred in women planning pregnancy. After inadvertent exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, close monitoring is necessary in pregnant women with supraventricular arrhythmias or cardiac disease.
Introduction
Ivabradine is a heart rate-lowering agent, acting through selective inhibition of the If (I-funny) channel without additional hemodynamic effects on cardiac function. The approved treatment indications in Europe are chronic heart failure and symptomatic treatment of chronic stable angina pectoris after failing or intolerance of beta-blockers [1].
Preclinical studies have suggested fetotoxicity and teratogenic effects. In pregnant rats, ivabradine exposure equivalent to therapeutic levels in humans was associated with increased intrauterine and postnatal mortality. Ventricular septal defects and complex anomalies of the great arteries were noted at doses 3 times the therapeutic human exposure. In pregnant rabbits, reduced fetal and placental weight as well as ectrodactyly was observed at doses 15 times the therapeutic human exposure [2]. In addition, dose-dependent mortality was seen in chicken and mice embryos [3, 4].
ESC guidelines consider ivabradine contraindicated during pregnancy and recommend discontinuation independent of treatment indication before conception with close clinical and echocardiographic monitoring [5].
Published data on ivabradine-exposed pregnancies are limited to three case reports, one with ivabradine treatment in the late first and early second trimester for sinus tachycardia after myocardial infarction [6], and the second starting ivabradine in the second trimester for deteriorating tachycardia-induced cardiomyopathy [7]. In the third case with first trimester exposure, pregnancy was electively terminated [8].
The limited human experience along with the teratogenic effects in animal studies makes it difficult to counsel inadvertently exposed pregnant women and their health care providers (HCP). Therefore, we decided to evaluate ivabradine-exposed pregnancies recorded by the German Embryotox Institute.
Methods
Study design
The German Embryotox Institute offers risk assessment on drug exposure during pregnancy to HCPs and patients. Up to 15,000 annual requests are answered, and in approximately 3500 critically exposed cases per year, the course and outcome of pregnancy are documented. At the initial contact, maternal characteristics and detailed exposure assessment are asked for after informed consent. About 8 weeks after the estimated date of birth, information on course and outcome of pregnancy is collected via structured telephone interview and mailed questionnaires. After a case by case plausibility check, additional information and health care records are requested in cases of adverse outcomes or inconsistent information.
Weeks of pregnancy were based on ultrasound and if not available on the last menstrual period (LMP). First trimester was defined as gestational week 2 + 0 (conception) until 12 + 6. Major birth defects were classified according to EUROCAT [9]. In prospectively ascertained pregnancies, neither the outcome of pregnancy nor prenatal pathology was known at the time of first contact. Pregnancies that were reported after birth or because of prenatal pathology were considered retrospective and evaluated separately. A detailed description of methodology adapted to the recommendations of the Strengthening of the Reporting of Observational studies in Epidemiology (STROBE) statement is given in Schaefer et al. [10] and Dathe and Schaefer [11].
We evaluated all requests on ivabradine to the German Embryotox Institute from 2007 to 2019.
Statistical analyses
Descriptive statistics were applied. For age, BMI, gestational weeks, neonatal weight, length, and head circumference, median and interquartile ranges are presented. Birth weights were adjusted to sex and gestational age at birth and percentile categories and standard deviation scores (SDS) were calculated according to the German perinatal survey [12].
Results
During the study period 2007–2019, our institute received 97 requests for ivabradine, 56 (58%) were from HCP, and 41 from patients. For further details, see Fig. 1. Treatment indications for ivabradine over time in relation to formal approval status and pregnancy labelling are summarized in Fig. 2. Supraventricular tachycardia was the most common reason for ivabradine treatment in our cohort. The first ivabradine-exposed pregnancy was recorded in March 2007, 1 year after approval and marketing authorization in Germany in January 2006.Fig. 1 Overview on information requests to the German Embryotox institute on ivabradine
Fig. 2 Treatment indication for ivabradine in 97 Embryotox requests over time in relation to pregnancy labelling and approval status by EMA/FDA. Recommendations for use of ivabradine in supraventricular tachycardias, i.e. inappropriate sinus tachycardia and postural orthostatic tachycardia syndrome: HRS/ESC 2015 [13], ACC/AHA/HFSA 2016 [14], ESC 2019 [15]. ACC, American College of Cardiology; EMA, European Medicines Agency; ESC, European Society of Cardiology; FDA, US Food and Drug Administration; HRS, Heart Rhythm Society
Fifty-six of the 97 requests were related to maternal exposure during pregnancy (Fig. 1). In 41 of these cases, follow-up on pregnancy outcome could be completed: 38 pregnancies were prospectively ascertained and three were retrospective reports.
Evaluation of prospectively ascertained pregnancies
Maternal characteristics
The median age of pregnant women was 29.5 years and the median BMI 22.5. Maternal characteristics are presented in detail in Table 1. The mean time at first contact was gestational week (GW) 7. Only 16 of the 38 prospectively ascertained pregnancies were actually planned (42%) and only 7 women (18%) started folic acid before conception. Contraceptive methods, mostly oral contraceptives (n = 4), were still used by 6 women at conception.Table 1 Maternal characteristics and pregnancy outcomes of ivabradine exposed pregnancies. Data are n (%) except for age, BMI, gestational week at birth, neonatal weight, length and head circumference, which are median and interquartile range
Maternal characteristics N (%) or median (IQR)
Maternal characteristics Age (n=38) 29.5 (26.5-34)
BMI (n=34) 22.5 (19.4-25.6)
Smoking (n=37)
>5 cigarettes/day 7 (19%)
<=5 cigarettes/day 4 (11%)
Alcohol (n=36)
<=1 drink/day 2 (6%)
>1 drink/day 1 (3%)
GW at first contact (n=38) 7 (5.6-11.1)
Previous pregnancies (n=38)
0 17(45%)
1 8 (21%)
2 13 (34%)
Pregnancy outcome and neonatal characteristics N (%) or median (IQR)
Pregnancy outcome (n=38) Spontaneous abortion 3
ETOP 3
Live birth 32
Pregnancy complications (n=32, only live births) Gestational diabetes 5 (16%)
Pre-eclampsia 2 (6%)
Cesarean section 15 (46%)
Preterm birth (<GW 37) 3 (9%)
Neonatal characteristics (n=33, including one pair of twins) GW at birth (n=32) 39 (37.9-39.9)
Weight, g (n=32) 3300 (2935-3610)
Length, cm (n=31) 50 (49-52.5)
HC, cm (n=30) 35 (34-36)
SGA (n=32) 2 (6%)
BMI, Body Mass Index; ETOP, elective termination of pregnancy; GW, gestational week; HC, head circumference; IQR, interquartile range; SGA, small for gestational age (<10th percentile)
Ivabradine exposure and co-medication
The most prevalent indication for ivabradine treatment was supraventricular tachycardia (n = 29) followed by heart failure (n = 3) and coronary artery disease (n = 2). Other reported indications (n = 4) included QT prolongation and sick-sinus-syndrome, although both are regarded as contraindications for ivabradine. The median daily ivabradine dose was 7.5 mg (IQR 5–10, min–max 2.5–15, n = 33).
Nearly all women were treated with ivabradine during the first trimester (n = 37) and 32 stopped treatment after recognition of pregnancy (Fig. 3), at median gestational week 6.1 (interquartile range 5.4–7.8). Cardiac decompensation or complications were not described after cessation of treatment. Only 5 women continued ivabradine throughout pregnancy or until shortly before delivery. One woman started treatment after first trimester (GW 18) of her twin pregnancy due to symptomatic tachycardia under beta-blocker therapy (Iv34 in Fig. 3).Fig. 3 Ivabradine exposure and pregnancy outcome of prospectively (n = 38) and retrospectively (n = 3) ascertained pregnancies
Cardiovascular co-medication consisted of beta-blockers (mainly metoprolol) in 15 women, mostly concomitantly (n = 7) or in replacement of ivabradine (n = 6). Two women initiated ivabradine after discontinuation of beta-blockers (Iv30 and Iv34 in Fig. 3). Six women received other cardiovascular medication, three methyldopa or amlodipine for hypertension, two digitoxin or verapamil for tachyarrhythmia, and one sildenafil and eplerenone for cardiomyopathy with cardiac failure (Iv17).
Psychiatric co-medication was reported in 7/38 women (5 of these with an anxiety disorder) including the teratogenic antiepileptic topiramate (n = 1) and pregabalin (n = 3).
Pregnancy outcome
Of the 38 prospectively ascertained ivabradine-exposed pregnancies, 32 were live births (33 infants including one pair of twins) and three resulted in early spontaneous abortions (Iv10, Iv17, and Iv19 in Fig. 3). One of these occurred in a woman with cardiomyopathy (Iv17) and various co-medications including hydroxycarbamide for thrombocythemia until GW 7. Three pregnancies were electively terminated for social/psychological reasons. Further information on pregnancy complications and outcomes is given in Table 1 and Fig. 3.
Major birth defects were reported in one pregnancy (Iv24) with 15 mg ivabradine plus bisoprolol until GW 7 + 5, both replaced by metoprolol. Polyhydramnios and preeclampsia led to Caesarean section in GW 35 after premature rupture of membranes. The premature child was large for gestational age (> 97 percentile) and had a cleft palate, an atrial septal defect II (spontaneously closed at the age of 10 months), and a hypertrophic cardiac septum. The mother was treated for sinus tachycardia, in addition she had diabetes type I and a BMI of 30.
Neonatal characteristics
Characteristics of the 33 neonates are summarized in Table 1. Compared to the German perinatal survey, standardized birth weights were lower with a median SDS of − 0.36, corresponding to a reduction of birth weight of 149 g for singletons born in gestational week 40.
Two neonates were small for gestational age, one of the twins (exposure only during 2nd and 3rd trimester, Iv34) and one after maternal ivabradine exposure until GW 12 (Iv29). Two neonates were large for gestational age, both mothers were diabetic (Iv24, type I diabetes and Iv27, gestational diabetes).
Retrospective cases
During the study period three pregnancies were retrospectively reported (further details are given in Table 2 and Fig. 3). One premature neonate (IvR1) with tracheal atresia died perinatally. The mother had been treated with ivabradine, naproxen, and metamizole during the first trimester until recognition of pregnancy (no further details available).Table 2 Retrospectively reported pregnancies
Case Maternal age/BMI Treatment indication Ivabradine exposure, other co-medication (GW or trimester) Pregnancy outcome, GW, birth weight percentile, sex Pregnancy complications, birth defects, and additional findings
IvR1 31 years/31.2 Tachycardia Ivabradine (1st trim)
Naproxen (1st trim)
Metamizole (1st trim)
Live birth, 36 + 2, > 97th, f Polyhydramnios, perinatal death, tracheal atresia, normal karyotype (46, XX)
IvR2 29 years/18.4 Tachycardia, syncopes, pacemaker Ivabradine (1st trim)
Bisoprolol (0–35 + 5)
Levetiracetam (0–35 + 5, grand mal epilepsy)
Live birth, 35 + 5, < 10th, f IUGR (since GW 30), oligohydramnios (since GW 34), Caesarean section (vaginal bleeding)
IvR3 29 years/not available Tachycardia Ivabradine (0–41 + 3)
Bisoprolol (0–41 + 3)
Live birth, 41 + 3
> 10th, m
Uncomplicated pregnancy
BMI, body mass index; f, female; GW, gestational week; IUGR, intrauterine growth restriction; m, male; trim, trimester
The other two retrospective cases include a woman with comorbidities, intrauterine growth restriction, and premature birth (IvR2). The second was a healthy newborn (IvR3) after treatment with ivabradine throughout pregnancy. This mother reported a previous also uneventful pregnancy course under ivabradine.
Discussion
Pregnancy outcome
In our prospective cohort, there was only one neonate with major birth defects (cleft palate and atrial septal defect) from a diabetic mother with supraventricular tachycardia. No further major or cardiac malformations were reported among the 32 live births and 6 pregnancy losses. Cardiac defects have been observed in preclinical experimental studies. As a possible teratogenic mechanism, ivabradine-induced embryonic heart rate reduction leading to reduced cardiac output and hypoxia was discussed [16]. With a prevalence of almost 1 of 100 cardiac malformations belong to the most frequent birth defects [17, 18]. In our case, the neonate was large for gestational age and had a septum hypertrophy, supporting the assumption that maternal pre-gestational diabetes may also be causative for the observed cardiac defect and cleft palate. A causal relationship between ivabradine and cleft palate is unlikely, because ivabradine was discontinued at GW 7 + 5 before the beginning of the critical time period for the development of cleft palate at gestational week 9. The rate of major birth defects in our prospective cohort is comparable to the expected background rate of 2–3% [17, 18]. However, the limited number of pregnancies exposed during the entire period of embryogenesis does not allow excluding an increased risk, so far.
There was only one retrospectively reported case with a birth defect (malformation of the trachea) suspected as potentially ivabradine related. The concomitant exposures with naproxen and metamizole have not been evidenced as teratogens in human, whereas the high maternal BMI in the present case poses a risk for birth defects.
Taken together, we did not observe a specific pattern of congenital anomalies in our study. In addition, the presence of maternal comorbidities like diabetes and high BMI may have contributed to the observed birth defects.
The number of spontaneous abortions (3/38) was not higher than expected. The rate of preeclampsia and preterm birth in our cohort was within the expected range [19].
Treatment indications
Although the approved treatment indications for ivabradine, chronic heart failure and chronic stable angina pectoris (see Fig. 2 for changes of licenced use during the study period), are rare in women of reproductive age, we have observed an increasing need for information for off-label use in (younger) women considering pregnancy.
In our case series, most women were treated (off-label) for supraventricular tachycardia using lower doses than recommended (5–7.5 mg twice daily). Supraventricular tachycardia is not uncommon, especially in younger women, and is expected in 0.02–0.5% of pregnancies [20]. Aggravation of symptoms during pregnancy is described in about 20–50% of women related to various mechanisms, including pregnancy-related increase in cardiac heart rate and output and hormonal changes [20].
In 2015, the HRS (Heart Rhythm Society) considered ivabradine as promising for treating patients with inappropriate sinus tachycardia (IST) [13], followed by a recommendation for ivabradine by the ACC (American College of Cardiology) in 2016 [14]—though only one small randomized controlled study has yet supported beneficial effects [21]. Although still not approved for this indication, recent ESC guidelines [15] also recommend ivabradine alone or in combination with a beta-blocker in symptomatic patients with IST, for the treatment of focal atrial tachycardia or for postural orthostatic tachycardia syndrome (POTS), a condition considered as common cause for orthostatic intolerance in women between 15 and 25 years [20, 22] (Fig. 2).
Beta-blockers were required in 18% of women in our prospective cohort (Fig. 3). Discontinuation of ivabradine or switching to beta-blockers after recognition of pregnancy was well tolerated in most of our patients. Thus, necessity of ivabradine treatment should be critically considered in women with supraventricular tachycardia planning pregnancy. Data from previous studies suggest that discontinuation of ivabradine is possible in many (non-pregnant) patients with IST without recurrence. Even beneficial long-term effects after cessation of therapy are discussed [23, 24].
A high number of women in our cohort received therapy for anxiety and/or other psychiatric conditions. This is in line with findings from other studies reporting high rates of anxiety (25%) and depression (26%) in IST patients [25]. Anxiety is considered as non-cardiac symptom of inappropriate sinus tachycardia, and vice versa, anxiety may trigger tachycardia. Especially during pregnancy, empathic care and reassurance are necessary, to prevent increased concerns about possible harm to the fetus [26].
Preconception counselling is of utmost importance in all women considering pregnancy under medication [27]. Sinus tachycardia is a condition diagnosed mainly in young females and the first diagnosis is frequently made during or after pregnancy [25]. Considering the increasing off-label use of ivabradine as well as the high rate of > 40% unplanned pregnancies [28], more exposed pregnancies can be expected in the future. Well-established treatment options such as selected beta-blockers with low reproductive risk [29] should be considered in women of childbearing age before prescription of ivabradine.
Strength and limitations
A major strength of our study is the detailed exposure protocol and the multi-source ascertainment of pregnancy outcome via patients and their HCP followed by a case by case plausibility test.
Our data may not be representative of the German pregnant population; insofar, women with higher education are over-represented [30]. However, the regional distribution of enrolled patients is representative of the female population of childbearing age in Germany [11]. In addition, women on long-term ivabradine with severe cardiovascular conditions such as the approved treatment indications coronary artery disease and cardiac failure are under-represented in our cohort possibly favouring uneventful pregnancy outcomes. However, given the rarity of severe cardiac conditions in pregnant women, large study cohorts are not to be expected in the foreseeable future.
Conclusions
Our findings do not indicate that ivabradine is a major teratogen when used in early pregnancy. Although being the largest study published so far on the outcome of ivabradine-exposed pregnancies, the sample is too small to rule out embryotoxic effects. Further studies are needed to confirm or refute our findings.
In women planning pregnancy, established drugs of choice with low risk for the unborn should be preferred to ivabradine. For women with significant arrhythmias and severe cardiac disease, the impact of the underlying medical condition including comorbidities such as diabetes and high BMI on pregnancy outcome has to be considered when planning pregnancy.
After inadvertent ivabradine exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, all pregnant women with complicated arrhythmias or major cardiac disease should be carefully monitored.
Acknowledgements
We would like to thank all patients and physicians reporting pregnancies to the institute and acknowledge our colleagues from the German Embryotox Institute for counselling and thoroughly documenting cases, especially Sandra Zinke for continuous support with case documentation.
Contributions of authors
MH, ML, and CS designed the study. ML, EB, MH, and KD validated and analysed the data. ML, MH, and CS wrote the first draft of the manuscript, and all the authors critically revised subsequent manuscript drafts and contributed essential discussion points. All the authors approved the final version of this manuscript and they are responsible for the accuracy of this work.
Funding
Open Access funding enabled and organized by Projekt DEAL. This work was performed with financial support from the German Ministry of Health (BMG) and the German Federal Institute for Drugs and Medical Devices (BfArM). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
The study was registered with the German Clinical Trial register (DRKS00010502) and the study protocol was approved by the ethics committee of the Charité-Universitätsmedizin Berlin (EA1/107/16).
The study was registered with the German Clinical Trial register (DRKS00010502, 20.5.2016)
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Recovered | ReactionOutcome | CC BY | 33501507 | 19,697,665 | 2021-07 |
What was the outcome of reaction 'Polyhydramnios'? | Ivabradine use in pregnant women-treatment indications and pregnancy outcome: an evaluation of the German Embryotox database.
OBJECTIVE
Ivabradine has been approved for the treatment of chronic heart failure and chronic stable angina pectoris in Europe. Based on adverse outcomes of reproductive animal studies and the lack of human data, ivabradine is considered contraindicated during pregnancy. The aim of this observational study is to analyse ivabradine use before and during pregnancy.
METHODS
We evaluated all ivabradine-related requests to the German Embryotox Institute from 2007 to 2019. Exposed pregnancies were analysed as to their outcome.
RESULTS
Off-label use for supraventricular tachycardia was frequent in women of childbearing age. Of 38 prospectively ascertained pregnancies with ivabradine exposure and completed follow-up, 32 resulted in live births, 3 in spontaneous abortions, and 3 were electively terminated. One neonate presented with major birth defects (atrial septal defect and cleft palate). In 33/38 patients, ivabradine was discontinued after confirmation of pregnancy without cardiac deterioration and 5/38 women continued ivabradine throughout pregnancy. In addition, there were 3 retrospectively reported pregnancies including one major birth defect (tracheal atresia).
CONCLUSIONS
This case series represents the largest cohort of ivabradine-exposed pregnancies, published so far. According to our findings, ivabradine appears not to be a major teratogen. However, established drugs of choice with strong evidence of low risk for the unborn should be preferred in women planning pregnancy. After inadvertent exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, close monitoring is necessary in pregnant women with supraventricular arrhythmias or cardiac disease.
Introduction
Ivabradine is a heart rate-lowering agent, acting through selective inhibition of the If (I-funny) channel without additional hemodynamic effects on cardiac function. The approved treatment indications in Europe are chronic heart failure and symptomatic treatment of chronic stable angina pectoris after failing or intolerance of beta-blockers [1].
Preclinical studies have suggested fetotoxicity and teratogenic effects. In pregnant rats, ivabradine exposure equivalent to therapeutic levels in humans was associated with increased intrauterine and postnatal mortality. Ventricular septal defects and complex anomalies of the great arteries were noted at doses 3 times the therapeutic human exposure. In pregnant rabbits, reduced fetal and placental weight as well as ectrodactyly was observed at doses 15 times the therapeutic human exposure [2]. In addition, dose-dependent mortality was seen in chicken and mice embryos [3, 4].
ESC guidelines consider ivabradine contraindicated during pregnancy and recommend discontinuation independent of treatment indication before conception with close clinical and echocardiographic monitoring [5].
Published data on ivabradine-exposed pregnancies are limited to three case reports, one with ivabradine treatment in the late first and early second trimester for sinus tachycardia after myocardial infarction [6], and the second starting ivabradine in the second trimester for deteriorating tachycardia-induced cardiomyopathy [7]. In the third case with first trimester exposure, pregnancy was electively terminated [8].
The limited human experience along with the teratogenic effects in animal studies makes it difficult to counsel inadvertently exposed pregnant women and their health care providers (HCP). Therefore, we decided to evaluate ivabradine-exposed pregnancies recorded by the German Embryotox Institute.
Methods
Study design
The German Embryotox Institute offers risk assessment on drug exposure during pregnancy to HCPs and patients. Up to 15,000 annual requests are answered, and in approximately 3500 critically exposed cases per year, the course and outcome of pregnancy are documented. At the initial contact, maternal characteristics and detailed exposure assessment are asked for after informed consent. About 8 weeks after the estimated date of birth, information on course and outcome of pregnancy is collected via structured telephone interview and mailed questionnaires. After a case by case plausibility check, additional information and health care records are requested in cases of adverse outcomes or inconsistent information.
Weeks of pregnancy were based on ultrasound and if not available on the last menstrual period (LMP). First trimester was defined as gestational week 2 + 0 (conception) until 12 + 6. Major birth defects were classified according to EUROCAT [9]. In prospectively ascertained pregnancies, neither the outcome of pregnancy nor prenatal pathology was known at the time of first contact. Pregnancies that were reported after birth or because of prenatal pathology were considered retrospective and evaluated separately. A detailed description of methodology adapted to the recommendations of the Strengthening of the Reporting of Observational studies in Epidemiology (STROBE) statement is given in Schaefer et al. [10] and Dathe and Schaefer [11].
We evaluated all requests on ivabradine to the German Embryotox Institute from 2007 to 2019.
Statistical analyses
Descriptive statistics were applied. For age, BMI, gestational weeks, neonatal weight, length, and head circumference, median and interquartile ranges are presented. Birth weights were adjusted to sex and gestational age at birth and percentile categories and standard deviation scores (SDS) were calculated according to the German perinatal survey [12].
Results
During the study period 2007–2019, our institute received 97 requests for ivabradine, 56 (58%) were from HCP, and 41 from patients. For further details, see Fig. 1. Treatment indications for ivabradine over time in relation to formal approval status and pregnancy labelling are summarized in Fig. 2. Supraventricular tachycardia was the most common reason for ivabradine treatment in our cohort. The first ivabradine-exposed pregnancy was recorded in March 2007, 1 year after approval and marketing authorization in Germany in January 2006.Fig. 1 Overview on information requests to the German Embryotox institute on ivabradine
Fig. 2 Treatment indication for ivabradine in 97 Embryotox requests over time in relation to pregnancy labelling and approval status by EMA/FDA. Recommendations for use of ivabradine in supraventricular tachycardias, i.e. inappropriate sinus tachycardia and postural orthostatic tachycardia syndrome: HRS/ESC 2015 [13], ACC/AHA/HFSA 2016 [14], ESC 2019 [15]. ACC, American College of Cardiology; EMA, European Medicines Agency; ESC, European Society of Cardiology; FDA, US Food and Drug Administration; HRS, Heart Rhythm Society
Fifty-six of the 97 requests were related to maternal exposure during pregnancy (Fig. 1). In 41 of these cases, follow-up on pregnancy outcome could be completed: 38 pregnancies were prospectively ascertained and three were retrospective reports.
Evaluation of prospectively ascertained pregnancies
Maternal characteristics
The median age of pregnant women was 29.5 years and the median BMI 22.5. Maternal characteristics are presented in detail in Table 1. The mean time at first contact was gestational week (GW) 7. Only 16 of the 38 prospectively ascertained pregnancies were actually planned (42%) and only 7 women (18%) started folic acid before conception. Contraceptive methods, mostly oral contraceptives (n = 4), were still used by 6 women at conception.Table 1 Maternal characteristics and pregnancy outcomes of ivabradine exposed pregnancies. Data are n (%) except for age, BMI, gestational week at birth, neonatal weight, length and head circumference, which are median and interquartile range
Maternal characteristics N (%) or median (IQR)
Maternal characteristics Age (n=38) 29.5 (26.5-34)
BMI (n=34) 22.5 (19.4-25.6)
Smoking (n=37)
>5 cigarettes/day 7 (19%)
<=5 cigarettes/day 4 (11%)
Alcohol (n=36)
<=1 drink/day 2 (6%)
>1 drink/day 1 (3%)
GW at first contact (n=38) 7 (5.6-11.1)
Previous pregnancies (n=38)
0 17(45%)
1 8 (21%)
2 13 (34%)
Pregnancy outcome and neonatal characteristics N (%) or median (IQR)
Pregnancy outcome (n=38) Spontaneous abortion 3
ETOP 3
Live birth 32
Pregnancy complications (n=32, only live births) Gestational diabetes 5 (16%)
Pre-eclampsia 2 (6%)
Cesarean section 15 (46%)
Preterm birth (<GW 37) 3 (9%)
Neonatal characteristics (n=33, including one pair of twins) GW at birth (n=32) 39 (37.9-39.9)
Weight, g (n=32) 3300 (2935-3610)
Length, cm (n=31) 50 (49-52.5)
HC, cm (n=30) 35 (34-36)
SGA (n=32) 2 (6%)
BMI, Body Mass Index; ETOP, elective termination of pregnancy; GW, gestational week; HC, head circumference; IQR, interquartile range; SGA, small for gestational age (<10th percentile)
Ivabradine exposure and co-medication
The most prevalent indication for ivabradine treatment was supraventricular tachycardia (n = 29) followed by heart failure (n = 3) and coronary artery disease (n = 2). Other reported indications (n = 4) included QT prolongation and sick-sinus-syndrome, although both are regarded as contraindications for ivabradine. The median daily ivabradine dose was 7.5 mg (IQR 5–10, min–max 2.5–15, n = 33).
Nearly all women were treated with ivabradine during the first trimester (n = 37) and 32 stopped treatment after recognition of pregnancy (Fig. 3), at median gestational week 6.1 (interquartile range 5.4–7.8). Cardiac decompensation or complications were not described after cessation of treatment. Only 5 women continued ivabradine throughout pregnancy or until shortly before delivery. One woman started treatment after first trimester (GW 18) of her twin pregnancy due to symptomatic tachycardia under beta-blocker therapy (Iv34 in Fig. 3).Fig. 3 Ivabradine exposure and pregnancy outcome of prospectively (n = 38) and retrospectively (n = 3) ascertained pregnancies
Cardiovascular co-medication consisted of beta-blockers (mainly metoprolol) in 15 women, mostly concomitantly (n = 7) or in replacement of ivabradine (n = 6). Two women initiated ivabradine after discontinuation of beta-blockers (Iv30 and Iv34 in Fig. 3). Six women received other cardiovascular medication, three methyldopa or amlodipine for hypertension, two digitoxin or verapamil for tachyarrhythmia, and one sildenafil and eplerenone for cardiomyopathy with cardiac failure (Iv17).
Psychiatric co-medication was reported in 7/38 women (5 of these with an anxiety disorder) including the teratogenic antiepileptic topiramate (n = 1) and pregabalin (n = 3).
Pregnancy outcome
Of the 38 prospectively ascertained ivabradine-exposed pregnancies, 32 were live births (33 infants including one pair of twins) and three resulted in early spontaneous abortions (Iv10, Iv17, and Iv19 in Fig. 3). One of these occurred in a woman with cardiomyopathy (Iv17) and various co-medications including hydroxycarbamide for thrombocythemia until GW 7. Three pregnancies were electively terminated for social/psychological reasons. Further information on pregnancy complications and outcomes is given in Table 1 and Fig. 3.
Major birth defects were reported in one pregnancy (Iv24) with 15 mg ivabradine plus bisoprolol until GW 7 + 5, both replaced by metoprolol. Polyhydramnios and preeclampsia led to Caesarean section in GW 35 after premature rupture of membranes. The premature child was large for gestational age (> 97 percentile) and had a cleft palate, an atrial septal defect II (spontaneously closed at the age of 10 months), and a hypertrophic cardiac septum. The mother was treated for sinus tachycardia, in addition she had diabetes type I and a BMI of 30.
Neonatal characteristics
Characteristics of the 33 neonates are summarized in Table 1. Compared to the German perinatal survey, standardized birth weights were lower with a median SDS of − 0.36, corresponding to a reduction of birth weight of 149 g for singletons born in gestational week 40.
Two neonates were small for gestational age, one of the twins (exposure only during 2nd and 3rd trimester, Iv34) and one after maternal ivabradine exposure until GW 12 (Iv29). Two neonates were large for gestational age, both mothers were diabetic (Iv24, type I diabetes and Iv27, gestational diabetes).
Retrospective cases
During the study period three pregnancies were retrospectively reported (further details are given in Table 2 and Fig. 3). One premature neonate (IvR1) with tracheal atresia died perinatally. The mother had been treated with ivabradine, naproxen, and metamizole during the first trimester until recognition of pregnancy (no further details available).Table 2 Retrospectively reported pregnancies
Case Maternal age/BMI Treatment indication Ivabradine exposure, other co-medication (GW or trimester) Pregnancy outcome, GW, birth weight percentile, sex Pregnancy complications, birth defects, and additional findings
IvR1 31 years/31.2 Tachycardia Ivabradine (1st trim)
Naproxen (1st trim)
Metamizole (1st trim)
Live birth, 36 + 2, > 97th, f Polyhydramnios, perinatal death, tracheal atresia, normal karyotype (46, XX)
IvR2 29 years/18.4 Tachycardia, syncopes, pacemaker Ivabradine (1st trim)
Bisoprolol (0–35 + 5)
Levetiracetam (0–35 + 5, grand mal epilepsy)
Live birth, 35 + 5, < 10th, f IUGR (since GW 30), oligohydramnios (since GW 34), Caesarean section (vaginal bleeding)
IvR3 29 years/not available Tachycardia Ivabradine (0–41 + 3)
Bisoprolol (0–41 + 3)
Live birth, 41 + 3
> 10th, m
Uncomplicated pregnancy
BMI, body mass index; f, female; GW, gestational week; IUGR, intrauterine growth restriction; m, male; trim, trimester
The other two retrospective cases include a woman with comorbidities, intrauterine growth restriction, and premature birth (IvR2). The second was a healthy newborn (IvR3) after treatment with ivabradine throughout pregnancy. This mother reported a previous also uneventful pregnancy course under ivabradine.
Discussion
Pregnancy outcome
In our prospective cohort, there was only one neonate with major birth defects (cleft palate and atrial septal defect) from a diabetic mother with supraventricular tachycardia. No further major or cardiac malformations were reported among the 32 live births and 6 pregnancy losses. Cardiac defects have been observed in preclinical experimental studies. As a possible teratogenic mechanism, ivabradine-induced embryonic heart rate reduction leading to reduced cardiac output and hypoxia was discussed [16]. With a prevalence of almost 1 of 100 cardiac malformations belong to the most frequent birth defects [17, 18]. In our case, the neonate was large for gestational age and had a septum hypertrophy, supporting the assumption that maternal pre-gestational diabetes may also be causative for the observed cardiac defect and cleft palate. A causal relationship between ivabradine and cleft palate is unlikely, because ivabradine was discontinued at GW 7 + 5 before the beginning of the critical time period for the development of cleft palate at gestational week 9. The rate of major birth defects in our prospective cohort is comparable to the expected background rate of 2–3% [17, 18]. However, the limited number of pregnancies exposed during the entire period of embryogenesis does not allow excluding an increased risk, so far.
There was only one retrospectively reported case with a birth defect (malformation of the trachea) suspected as potentially ivabradine related. The concomitant exposures with naproxen and metamizole have not been evidenced as teratogens in human, whereas the high maternal BMI in the present case poses a risk for birth defects.
Taken together, we did not observe a specific pattern of congenital anomalies in our study. In addition, the presence of maternal comorbidities like diabetes and high BMI may have contributed to the observed birth defects.
The number of spontaneous abortions (3/38) was not higher than expected. The rate of preeclampsia and preterm birth in our cohort was within the expected range [19].
Treatment indications
Although the approved treatment indications for ivabradine, chronic heart failure and chronic stable angina pectoris (see Fig. 2 for changes of licenced use during the study period), are rare in women of reproductive age, we have observed an increasing need for information for off-label use in (younger) women considering pregnancy.
In our case series, most women were treated (off-label) for supraventricular tachycardia using lower doses than recommended (5–7.5 mg twice daily). Supraventricular tachycardia is not uncommon, especially in younger women, and is expected in 0.02–0.5% of pregnancies [20]. Aggravation of symptoms during pregnancy is described in about 20–50% of women related to various mechanisms, including pregnancy-related increase in cardiac heart rate and output and hormonal changes [20].
In 2015, the HRS (Heart Rhythm Society) considered ivabradine as promising for treating patients with inappropriate sinus tachycardia (IST) [13], followed by a recommendation for ivabradine by the ACC (American College of Cardiology) in 2016 [14]—though only one small randomized controlled study has yet supported beneficial effects [21]. Although still not approved for this indication, recent ESC guidelines [15] also recommend ivabradine alone or in combination with a beta-blocker in symptomatic patients with IST, for the treatment of focal atrial tachycardia or for postural orthostatic tachycardia syndrome (POTS), a condition considered as common cause for orthostatic intolerance in women between 15 and 25 years [20, 22] (Fig. 2).
Beta-blockers were required in 18% of women in our prospective cohort (Fig. 3). Discontinuation of ivabradine or switching to beta-blockers after recognition of pregnancy was well tolerated in most of our patients. Thus, necessity of ivabradine treatment should be critically considered in women with supraventricular tachycardia planning pregnancy. Data from previous studies suggest that discontinuation of ivabradine is possible in many (non-pregnant) patients with IST without recurrence. Even beneficial long-term effects after cessation of therapy are discussed [23, 24].
A high number of women in our cohort received therapy for anxiety and/or other psychiatric conditions. This is in line with findings from other studies reporting high rates of anxiety (25%) and depression (26%) in IST patients [25]. Anxiety is considered as non-cardiac symptom of inappropriate sinus tachycardia, and vice versa, anxiety may trigger tachycardia. Especially during pregnancy, empathic care and reassurance are necessary, to prevent increased concerns about possible harm to the fetus [26].
Preconception counselling is of utmost importance in all women considering pregnancy under medication [27]. Sinus tachycardia is a condition diagnosed mainly in young females and the first diagnosis is frequently made during or after pregnancy [25]. Considering the increasing off-label use of ivabradine as well as the high rate of > 40% unplanned pregnancies [28], more exposed pregnancies can be expected in the future. Well-established treatment options such as selected beta-blockers with low reproductive risk [29] should be considered in women of childbearing age before prescription of ivabradine.
Strength and limitations
A major strength of our study is the detailed exposure protocol and the multi-source ascertainment of pregnancy outcome via patients and their HCP followed by a case by case plausibility test.
Our data may not be representative of the German pregnant population; insofar, women with higher education are over-represented [30]. However, the regional distribution of enrolled patients is representative of the female population of childbearing age in Germany [11]. In addition, women on long-term ivabradine with severe cardiovascular conditions such as the approved treatment indications coronary artery disease and cardiac failure are under-represented in our cohort possibly favouring uneventful pregnancy outcomes. However, given the rarity of severe cardiac conditions in pregnant women, large study cohorts are not to be expected in the foreseeable future.
Conclusions
Our findings do not indicate that ivabradine is a major teratogen when used in early pregnancy. Although being the largest study published so far on the outcome of ivabradine-exposed pregnancies, the sample is too small to rule out embryotoxic effects. Further studies are needed to confirm or refute our findings.
In women planning pregnancy, established drugs of choice with low risk for the unborn should be preferred to ivabradine. For women with significant arrhythmias and severe cardiac disease, the impact of the underlying medical condition including comorbidities such as diabetes and high BMI on pregnancy outcome has to be considered when planning pregnancy.
After inadvertent ivabradine exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, all pregnant women with complicated arrhythmias or major cardiac disease should be carefully monitored.
Acknowledgements
We would like to thank all patients and physicians reporting pregnancies to the institute and acknowledge our colleagues from the German Embryotox Institute for counselling and thoroughly documenting cases, especially Sandra Zinke for continuous support with case documentation.
Contributions of authors
MH, ML, and CS designed the study. ML, EB, MH, and KD validated and analysed the data. ML, MH, and CS wrote the first draft of the manuscript, and all the authors critically revised subsequent manuscript drafts and contributed essential discussion points. All the authors approved the final version of this manuscript and they are responsible for the accuracy of this work.
Funding
Open Access funding enabled and organized by Projekt DEAL. This work was performed with financial support from the German Ministry of Health (BMG) and the German Federal Institute for Drugs and Medical Devices (BfArM). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
The study was registered with the German Clinical Trial register (DRKS00010502) and the study protocol was approved by the ethics committee of the Charité-Universitätsmedizin Berlin (EA1/107/16).
The study was registered with the German Clinical Trial register (DRKS00010502, 20.5.2016)
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Recovered | ReactionOutcome | CC BY | 33501507 | 19,157,674 | 2021-07 |
What was the outcome of reaction 'Pre-eclampsia'? | Ivabradine use in pregnant women-treatment indications and pregnancy outcome: an evaluation of the German Embryotox database.
OBJECTIVE
Ivabradine has been approved for the treatment of chronic heart failure and chronic stable angina pectoris in Europe. Based on adverse outcomes of reproductive animal studies and the lack of human data, ivabradine is considered contraindicated during pregnancy. The aim of this observational study is to analyse ivabradine use before and during pregnancy.
METHODS
We evaluated all ivabradine-related requests to the German Embryotox Institute from 2007 to 2019. Exposed pregnancies were analysed as to their outcome.
RESULTS
Off-label use for supraventricular tachycardia was frequent in women of childbearing age. Of 38 prospectively ascertained pregnancies with ivabradine exposure and completed follow-up, 32 resulted in live births, 3 in spontaneous abortions, and 3 were electively terminated. One neonate presented with major birth defects (atrial septal defect and cleft palate). In 33/38 patients, ivabradine was discontinued after confirmation of pregnancy without cardiac deterioration and 5/38 women continued ivabradine throughout pregnancy. In addition, there were 3 retrospectively reported pregnancies including one major birth defect (tracheal atresia).
CONCLUSIONS
This case series represents the largest cohort of ivabradine-exposed pregnancies, published so far. According to our findings, ivabradine appears not to be a major teratogen. However, established drugs of choice with strong evidence of low risk for the unborn should be preferred in women planning pregnancy. After inadvertent exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, close monitoring is necessary in pregnant women with supraventricular arrhythmias or cardiac disease.
Introduction
Ivabradine is a heart rate-lowering agent, acting through selective inhibition of the If (I-funny) channel without additional hemodynamic effects on cardiac function. The approved treatment indications in Europe are chronic heart failure and symptomatic treatment of chronic stable angina pectoris after failing or intolerance of beta-blockers [1].
Preclinical studies have suggested fetotoxicity and teratogenic effects. In pregnant rats, ivabradine exposure equivalent to therapeutic levels in humans was associated with increased intrauterine and postnatal mortality. Ventricular septal defects and complex anomalies of the great arteries were noted at doses 3 times the therapeutic human exposure. In pregnant rabbits, reduced fetal and placental weight as well as ectrodactyly was observed at doses 15 times the therapeutic human exposure [2]. In addition, dose-dependent mortality was seen in chicken and mice embryos [3, 4].
ESC guidelines consider ivabradine contraindicated during pregnancy and recommend discontinuation independent of treatment indication before conception with close clinical and echocardiographic monitoring [5].
Published data on ivabradine-exposed pregnancies are limited to three case reports, one with ivabradine treatment in the late first and early second trimester for sinus tachycardia after myocardial infarction [6], and the second starting ivabradine in the second trimester for deteriorating tachycardia-induced cardiomyopathy [7]. In the third case with first trimester exposure, pregnancy was electively terminated [8].
The limited human experience along with the teratogenic effects in animal studies makes it difficult to counsel inadvertently exposed pregnant women and their health care providers (HCP). Therefore, we decided to evaluate ivabradine-exposed pregnancies recorded by the German Embryotox Institute.
Methods
Study design
The German Embryotox Institute offers risk assessment on drug exposure during pregnancy to HCPs and patients. Up to 15,000 annual requests are answered, and in approximately 3500 critically exposed cases per year, the course and outcome of pregnancy are documented. At the initial contact, maternal characteristics and detailed exposure assessment are asked for after informed consent. About 8 weeks after the estimated date of birth, information on course and outcome of pregnancy is collected via structured telephone interview and mailed questionnaires. After a case by case plausibility check, additional information and health care records are requested in cases of adverse outcomes or inconsistent information.
Weeks of pregnancy were based on ultrasound and if not available on the last menstrual period (LMP). First trimester was defined as gestational week 2 + 0 (conception) until 12 + 6. Major birth defects were classified according to EUROCAT [9]. In prospectively ascertained pregnancies, neither the outcome of pregnancy nor prenatal pathology was known at the time of first contact. Pregnancies that were reported after birth or because of prenatal pathology were considered retrospective and evaluated separately. A detailed description of methodology adapted to the recommendations of the Strengthening of the Reporting of Observational studies in Epidemiology (STROBE) statement is given in Schaefer et al. [10] and Dathe and Schaefer [11].
We evaluated all requests on ivabradine to the German Embryotox Institute from 2007 to 2019.
Statistical analyses
Descriptive statistics were applied. For age, BMI, gestational weeks, neonatal weight, length, and head circumference, median and interquartile ranges are presented. Birth weights were adjusted to sex and gestational age at birth and percentile categories and standard deviation scores (SDS) were calculated according to the German perinatal survey [12].
Results
During the study period 2007–2019, our institute received 97 requests for ivabradine, 56 (58%) were from HCP, and 41 from patients. For further details, see Fig. 1. Treatment indications for ivabradine over time in relation to formal approval status and pregnancy labelling are summarized in Fig. 2. Supraventricular tachycardia was the most common reason for ivabradine treatment in our cohort. The first ivabradine-exposed pregnancy was recorded in March 2007, 1 year after approval and marketing authorization in Germany in January 2006.Fig. 1 Overview on information requests to the German Embryotox institute on ivabradine
Fig. 2 Treatment indication for ivabradine in 97 Embryotox requests over time in relation to pregnancy labelling and approval status by EMA/FDA. Recommendations for use of ivabradine in supraventricular tachycardias, i.e. inappropriate sinus tachycardia and postural orthostatic tachycardia syndrome: HRS/ESC 2015 [13], ACC/AHA/HFSA 2016 [14], ESC 2019 [15]. ACC, American College of Cardiology; EMA, European Medicines Agency; ESC, European Society of Cardiology; FDA, US Food and Drug Administration; HRS, Heart Rhythm Society
Fifty-six of the 97 requests were related to maternal exposure during pregnancy (Fig. 1). In 41 of these cases, follow-up on pregnancy outcome could be completed: 38 pregnancies were prospectively ascertained and three were retrospective reports.
Evaluation of prospectively ascertained pregnancies
Maternal characteristics
The median age of pregnant women was 29.5 years and the median BMI 22.5. Maternal characteristics are presented in detail in Table 1. The mean time at first contact was gestational week (GW) 7. Only 16 of the 38 prospectively ascertained pregnancies were actually planned (42%) and only 7 women (18%) started folic acid before conception. Contraceptive methods, mostly oral contraceptives (n = 4), were still used by 6 women at conception.Table 1 Maternal characteristics and pregnancy outcomes of ivabradine exposed pregnancies. Data are n (%) except for age, BMI, gestational week at birth, neonatal weight, length and head circumference, which are median and interquartile range
Maternal characteristics N (%) or median (IQR)
Maternal characteristics Age (n=38) 29.5 (26.5-34)
BMI (n=34) 22.5 (19.4-25.6)
Smoking (n=37)
>5 cigarettes/day 7 (19%)
<=5 cigarettes/day 4 (11%)
Alcohol (n=36)
<=1 drink/day 2 (6%)
>1 drink/day 1 (3%)
GW at first contact (n=38) 7 (5.6-11.1)
Previous pregnancies (n=38)
0 17(45%)
1 8 (21%)
2 13 (34%)
Pregnancy outcome and neonatal characteristics N (%) or median (IQR)
Pregnancy outcome (n=38) Spontaneous abortion 3
ETOP 3
Live birth 32
Pregnancy complications (n=32, only live births) Gestational diabetes 5 (16%)
Pre-eclampsia 2 (6%)
Cesarean section 15 (46%)
Preterm birth (<GW 37) 3 (9%)
Neonatal characteristics (n=33, including one pair of twins) GW at birth (n=32) 39 (37.9-39.9)
Weight, g (n=32) 3300 (2935-3610)
Length, cm (n=31) 50 (49-52.5)
HC, cm (n=30) 35 (34-36)
SGA (n=32) 2 (6%)
BMI, Body Mass Index; ETOP, elective termination of pregnancy; GW, gestational week; HC, head circumference; IQR, interquartile range; SGA, small for gestational age (<10th percentile)
Ivabradine exposure and co-medication
The most prevalent indication for ivabradine treatment was supraventricular tachycardia (n = 29) followed by heart failure (n = 3) and coronary artery disease (n = 2). Other reported indications (n = 4) included QT prolongation and sick-sinus-syndrome, although both are regarded as contraindications for ivabradine. The median daily ivabradine dose was 7.5 mg (IQR 5–10, min–max 2.5–15, n = 33).
Nearly all women were treated with ivabradine during the first trimester (n = 37) and 32 stopped treatment after recognition of pregnancy (Fig. 3), at median gestational week 6.1 (interquartile range 5.4–7.8). Cardiac decompensation or complications were not described after cessation of treatment. Only 5 women continued ivabradine throughout pregnancy or until shortly before delivery. One woman started treatment after first trimester (GW 18) of her twin pregnancy due to symptomatic tachycardia under beta-blocker therapy (Iv34 in Fig. 3).Fig. 3 Ivabradine exposure and pregnancy outcome of prospectively (n = 38) and retrospectively (n = 3) ascertained pregnancies
Cardiovascular co-medication consisted of beta-blockers (mainly metoprolol) in 15 women, mostly concomitantly (n = 7) or in replacement of ivabradine (n = 6). Two women initiated ivabradine after discontinuation of beta-blockers (Iv30 and Iv34 in Fig. 3). Six women received other cardiovascular medication, three methyldopa or amlodipine for hypertension, two digitoxin or verapamil for tachyarrhythmia, and one sildenafil and eplerenone for cardiomyopathy with cardiac failure (Iv17).
Psychiatric co-medication was reported in 7/38 women (5 of these with an anxiety disorder) including the teratogenic antiepileptic topiramate (n = 1) and pregabalin (n = 3).
Pregnancy outcome
Of the 38 prospectively ascertained ivabradine-exposed pregnancies, 32 were live births (33 infants including one pair of twins) and three resulted in early spontaneous abortions (Iv10, Iv17, and Iv19 in Fig. 3). One of these occurred in a woman with cardiomyopathy (Iv17) and various co-medications including hydroxycarbamide for thrombocythemia until GW 7. Three pregnancies were electively terminated for social/psychological reasons. Further information on pregnancy complications and outcomes is given in Table 1 and Fig. 3.
Major birth defects were reported in one pregnancy (Iv24) with 15 mg ivabradine plus bisoprolol until GW 7 + 5, both replaced by metoprolol. Polyhydramnios and preeclampsia led to Caesarean section in GW 35 after premature rupture of membranes. The premature child was large for gestational age (> 97 percentile) and had a cleft palate, an atrial septal defect II (spontaneously closed at the age of 10 months), and a hypertrophic cardiac septum. The mother was treated for sinus tachycardia, in addition she had diabetes type I and a BMI of 30.
Neonatal characteristics
Characteristics of the 33 neonates are summarized in Table 1. Compared to the German perinatal survey, standardized birth weights were lower with a median SDS of − 0.36, corresponding to a reduction of birth weight of 149 g for singletons born in gestational week 40.
Two neonates were small for gestational age, one of the twins (exposure only during 2nd and 3rd trimester, Iv34) and one after maternal ivabradine exposure until GW 12 (Iv29). Two neonates were large for gestational age, both mothers were diabetic (Iv24, type I diabetes and Iv27, gestational diabetes).
Retrospective cases
During the study period three pregnancies were retrospectively reported (further details are given in Table 2 and Fig. 3). One premature neonate (IvR1) with tracheal atresia died perinatally. The mother had been treated with ivabradine, naproxen, and metamizole during the first trimester until recognition of pregnancy (no further details available).Table 2 Retrospectively reported pregnancies
Case Maternal age/BMI Treatment indication Ivabradine exposure, other co-medication (GW or trimester) Pregnancy outcome, GW, birth weight percentile, sex Pregnancy complications, birth defects, and additional findings
IvR1 31 years/31.2 Tachycardia Ivabradine (1st trim)
Naproxen (1st trim)
Metamizole (1st trim)
Live birth, 36 + 2, > 97th, f Polyhydramnios, perinatal death, tracheal atresia, normal karyotype (46, XX)
IvR2 29 years/18.4 Tachycardia, syncopes, pacemaker Ivabradine (1st trim)
Bisoprolol (0–35 + 5)
Levetiracetam (0–35 + 5, grand mal epilepsy)
Live birth, 35 + 5, < 10th, f IUGR (since GW 30), oligohydramnios (since GW 34), Caesarean section (vaginal bleeding)
IvR3 29 years/not available Tachycardia Ivabradine (0–41 + 3)
Bisoprolol (0–41 + 3)
Live birth, 41 + 3
> 10th, m
Uncomplicated pregnancy
BMI, body mass index; f, female; GW, gestational week; IUGR, intrauterine growth restriction; m, male; trim, trimester
The other two retrospective cases include a woman with comorbidities, intrauterine growth restriction, and premature birth (IvR2). The second was a healthy newborn (IvR3) after treatment with ivabradine throughout pregnancy. This mother reported a previous also uneventful pregnancy course under ivabradine.
Discussion
Pregnancy outcome
In our prospective cohort, there was only one neonate with major birth defects (cleft palate and atrial septal defect) from a diabetic mother with supraventricular tachycardia. No further major or cardiac malformations were reported among the 32 live births and 6 pregnancy losses. Cardiac defects have been observed in preclinical experimental studies. As a possible teratogenic mechanism, ivabradine-induced embryonic heart rate reduction leading to reduced cardiac output and hypoxia was discussed [16]. With a prevalence of almost 1 of 100 cardiac malformations belong to the most frequent birth defects [17, 18]. In our case, the neonate was large for gestational age and had a septum hypertrophy, supporting the assumption that maternal pre-gestational diabetes may also be causative for the observed cardiac defect and cleft palate. A causal relationship between ivabradine and cleft palate is unlikely, because ivabradine was discontinued at GW 7 + 5 before the beginning of the critical time period for the development of cleft palate at gestational week 9. The rate of major birth defects in our prospective cohort is comparable to the expected background rate of 2–3% [17, 18]. However, the limited number of pregnancies exposed during the entire period of embryogenesis does not allow excluding an increased risk, so far.
There was only one retrospectively reported case with a birth defect (malformation of the trachea) suspected as potentially ivabradine related. The concomitant exposures with naproxen and metamizole have not been evidenced as teratogens in human, whereas the high maternal BMI in the present case poses a risk for birth defects.
Taken together, we did not observe a specific pattern of congenital anomalies in our study. In addition, the presence of maternal comorbidities like diabetes and high BMI may have contributed to the observed birth defects.
The number of spontaneous abortions (3/38) was not higher than expected. The rate of preeclampsia and preterm birth in our cohort was within the expected range [19].
Treatment indications
Although the approved treatment indications for ivabradine, chronic heart failure and chronic stable angina pectoris (see Fig. 2 for changes of licenced use during the study period), are rare in women of reproductive age, we have observed an increasing need for information for off-label use in (younger) women considering pregnancy.
In our case series, most women were treated (off-label) for supraventricular tachycardia using lower doses than recommended (5–7.5 mg twice daily). Supraventricular tachycardia is not uncommon, especially in younger women, and is expected in 0.02–0.5% of pregnancies [20]. Aggravation of symptoms during pregnancy is described in about 20–50% of women related to various mechanisms, including pregnancy-related increase in cardiac heart rate and output and hormonal changes [20].
In 2015, the HRS (Heart Rhythm Society) considered ivabradine as promising for treating patients with inappropriate sinus tachycardia (IST) [13], followed by a recommendation for ivabradine by the ACC (American College of Cardiology) in 2016 [14]—though only one small randomized controlled study has yet supported beneficial effects [21]. Although still not approved for this indication, recent ESC guidelines [15] also recommend ivabradine alone or in combination with a beta-blocker in symptomatic patients with IST, for the treatment of focal atrial tachycardia or for postural orthostatic tachycardia syndrome (POTS), a condition considered as common cause for orthostatic intolerance in women between 15 and 25 years [20, 22] (Fig. 2).
Beta-blockers were required in 18% of women in our prospective cohort (Fig. 3). Discontinuation of ivabradine or switching to beta-blockers after recognition of pregnancy was well tolerated in most of our patients. Thus, necessity of ivabradine treatment should be critically considered in women with supraventricular tachycardia planning pregnancy. Data from previous studies suggest that discontinuation of ivabradine is possible in many (non-pregnant) patients with IST without recurrence. Even beneficial long-term effects after cessation of therapy are discussed [23, 24].
A high number of women in our cohort received therapy for anxiety and/or other psychiatric conditions. This is in line with findings from other studies reporting high rates of anxiety (25%) and depression (26%) in IST patients [25]. Anxiety is considered as non-cardiac symptom of inappropriate sinus tachycardia, and vice versa, anxiety may trigger tachycardia. Especially during pregnancy, empathic care and reassurance are necessary, to prevent increased concerns about possible harm to the fetus [26].
Preconception counselling is of utmost importance in all women considering pregnancy under medication [27]. Sinus tachycardia is a condition diagnosed mainly in young females and the first diagnosis is frequently made during or after pregnancy [25]. Considering the increasing off-label use of ivabradine as well as the high rate of > 40% unplanned pregnancies [28], more exposed pregnancies can be expected in the future. Well-established treatment options such as selected beta-blockers with low reproductive risk [29] should be considered in women of childbearing age before prescription of ivabradine.
Strength and limitations
A major strength of our study is the detailed exposure protocol and the multi-source ascertainment of pregnancy outcome via patients and their HCP followed by a case by case plausibility test.
Our data may not be representative of the German pregnant population; insofar, women with higher education are over-represented [30]. However, the regional distribution of enrolled patients is representative of the female population of childbearing age in Germany [11]. In addition, women on long-term ivabradine with severe cardiovascular conditions such as the approved treatment indications coronary artery disease and cardiac failure are under-represented in our cohort possibly favouring uneventful pregnancy outcomes. However, given the rarity of severe cardiac conditions in pregnant women, large study cohorts are not to be expected in the foreseeable future.
Conclusions
Our findings do not indicate that ivabradine is a major teratogen when used in early pregnancy. Although being the largest study published so far on the outcome of ivabradine-exposed pregnancies, the sample is too small to rule out embryotoxic effects. Further studies are needed to confirm or refute our findings.
In women planning pregnancy, established drugs of choice with low risk for the unborn should be preferred to ivabradine. For women with significant arrhythmias and severe cardiac disease, the impact of the underlying medical condition including comorbidities such as diabetes and high BMI on pregnancy outcome has to be considered when planning pregnancy.
After inadvertent ivabradine exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, all pregnant women with complicated arrhythmias or major cardiac disease should be carefully monitored.
Acknowledgements
We would like to thank all patients and physicians reporting pregnancies to the institute and acknowledge our colleagues from the German Embryotox Institute for counselling and thoroughly documenting cases, especially Sandra Zinke for continuous support with case documentation.
Contributions of authors
MH, ML, and CS designed the study. ML, EB, MH, and KD validated and analysed the data. ML, MH, and CS wrote the first draft of the manuscript, and all the authors critically revised subsequent manuscript drafts and contributed essential discussion points. All the authors approved the final version of this manuscript and they are responsible for the accuracy of this work.
Funding
Open Access funding enabled and organized by Projekt DEAL. This work was performed with financial support from the German Ministry of Health (BMG) and the German Federal Institute for Drugs and Medical Devices (BfArM). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
The study was registered with the German Clinical Trial register (DRKS00010502) and the study protocol was approved by the ethics committee of the Charité-Universitätsmedizin Berlin (EA1/107/16).
The study was registered with the German Clinical Trial register (DRKS00010502, 20.5.2016)
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Recovered | ReactionOutcome | CC BY | 33501507 | 19,157,674 | 2021-07 |
What was the outcome of reaction 'Premature rupture of membranes'? | Ivabradine use in pregnant women-treatment indications and pregnancy outcome: an evaluation of the German Embryotox database.
OBJECTIVE
Ivabradine has been approved for the treatment of chronic heart failure and chronic stable angina pectoris in Europe. Based on adverse outcomes of reproductive animal studies and the lack of human data, ivabradine is considered contraindicated during pregnancy. The aim of this observational study is to analyse ivabradine use before and during pregnancy.
METHODS
We evaluated all ivabradine-related requests to the German Embryotox Institute from 2007 to 2019. Exposed pregnancies were analysed as to their outcome.
RESULTS
Off-label use for supraventricular tachycardia was frequent in women of childbearing age. Of 38 prospectively ascertained pregnancies with ivabradine exposure and completed follow-up, 32 resulted in live births, 3 in spontaneous abortions, and 3 were electively terminated. One neonate presented with major birth defects (atrial septal defect and cleft palate). In 33/38 patients, ivabradine was discontinued after confirmation of pregnancy without cardiac deterioration and 5/38 women continued ivabradine throughout pregnancy. In addition, there were 3 retrospectively reported pregnancies including one major birth defect (tracheal atresia).
CONCLUSIONS
This case series represents the largest cohort of ivabradine-exposed pregnancies, published so far. According to our findings, ivabradine appears not to be a major teratogen. However, established drugs of choice with strong evidence of low risk for the unborn should be preferred in women planning pregnancy. After inadvertent exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, close monitoring is necessary in pregnant women with supraventricular arrhythmias or cardiac disease.
Introduction
Ivabradine is a heart rate-lowering agent, acting through selective inhibition of the If (I-funny) channel without additional hemodynamic effects on cardiac function. The approved treatment indications in Europe are chronic heart failure and symptomatic treatment of chronic stable angina pectoris after failing or intolerance of beta-blockers [1].
Preclinical studies have suggested fetotoxicity and teratogenic effects. In pregnant rats, ivabradine exposure equivalent to therapeutic levels in humans was associated with increased intrauterine and postnatal mortality. Ventricular septal defects and complex anomalies of the great arteries were noted at doses 3 times the therapeutic human exposure. In pregnant rabbits, reduced fetal and placental weight as well as ectrodactyly was observed at doses 15 times the therapeutic human exposure [2]. In addition, dose-dependent mortality was seen in chicken and mice embryos [3, 4].
ESC guidelines consider ivabradine contraindicated during pregnancy and recommend discontinuation independent of treatment indication before conception with close clinical and echocardiographic monitoring [5].
Published data on ivabradine-exposed pregnancies are limited to three case reports, one with ivabradine treatment in the late first and early second trimester for sinus tachycardia after myocardial infarction [6], and the second starting ivabradine in the second trimester for deteriorating tachycardia-induced cardiomyopathy [7]. In the third case with first trimester exposure, pregnancy was electively terminated [8].
The limited human experience along with the teratogenic effects in animal studies makes it difficult to counsel inadvertently exposed pregnant women and their health care providers (HCP). Therefore, we decided to evaluate ivabradine-exposed pregnancies recorded by the German Embryotox Institute.
Methods
Study design
The German Embryotox Institute offers risk assessment on drug exposure during pregnancy to HCPs and patients. Up to 15,000 annual requests are answered, and in approximately 3500 critically exposed cases per year, the course and outcome of pregnancy are documented. At the initial contact, maternal characteristics and detailed exposure assessment are asked for after informed consent. About 8 weeks after the estimated date of birth, information on course and outcome of pregnancy is collected via structured telephone interview and mailed questionnaires. After a case by case plausibility check, additional information and health care records are requested in cases of adverse outcomes or inconsistent information.
Weeks of pregnancy were based on ultrasound and if not available on the last menstrual period (LMP). First trimester was defined as gestational week 2 + 0 (conception) until 12 + 6. Major birth defects were classified according to EUROCAT [9]. In prospectively ascertained pregnancies, neither the outcome of pregnancy nor prenatal pathology was known at the time of first contact. Pregnancies that were reported after birth or because of prenatal pathology were considered retrospective and evaluated separately. A detailed description of methodology adapted to the recommendations of the Strengthening of the Reporting of Observational studies in Epidemiology (STROBE) statement is given in Schaefer et al. [10] and Dathe and Schaefer [11].
We evaluated all requests on ivabradine to the German Embryotox Institute from 2007 to 2019.
Statistical analyses
Descriptive statistics were applied. For age, BMI, gestational weeks, neonatal weight, length, and head circumference, median and interquartile ranges are presented. Birth weights were adjusted to sex and gestational age at birth and percentile categories and standard deviation scores (SDS) were calculated according to the German perinatal survey [12].
Results
During the study period 2007–2019, our institute received 97 requests for ivabradine, 56 (58%) were from HCP, and 41 from patients. For further details, see Fig. 1. Treatment indications for ivabradine over time in relation to formal approval status and pregnancy labelling are summarized in Fig. 2. Supraventricular tachycardia was the most common reason for ivabradine treatment in our cohort. The first ivabradine-exposed pregnancy was recorded in March 2007, 1 year after approval and marketing authorization in Germany in January 2006.Fig. 1 Overview on information requests to the German Embryotox institute on ivabradine
Fig. 2 Treatment indication for ivabradine in 97 Embryotox requests over time in relation to pregnancy labelling and approval status by EMA/FDA. Recommendations for use of ivabradine in supraventricular tachycardias, i.e. inappropriate sinus tachycardia and postural orthostatic tachycardia syndrome: HRS/ESC 2015 [13], ACC/AHA/HFSA 2016 [14], ESC 2019 [15]. ACC, American College of Cardiology; EMA, European Medicines Agency; ESC, European Society of Cardiology; FDA, US Food and Drug Administration; HRS, Heart Rhythm Society
Fifty-six of the 97 requests were related to maternal exposure during pregnancy (Fig. 1). In 41 of these cases, follow-up on pregnancy outcome could be completed: 38 pregnancies were prospectively ascertained and three were retrospective reports.
Evaluation of prospectively ascertained pregnancies
Maternal characteristics
The median age of pregnant women was 29.5 years and the median BMI 22.5. Maternal characteristics are presented in detail in Table 1. The mean time at first contact was gestational week (GW) 7. Only 16 of the 38 prospectively ascertained pregnancies were actually planned (42%) and only 7 women (18%) started folic acid before conception. Contraceptive methods, mostly oral contraceptives (n = 4), were still used by 6 women at conception.Table 1 Maternal characteristics and pregnancy outcomes of ivabradine exposed pregnancies. Data are n (%) except for age, BMI, gestational week at birth, neonatal weight, length and head circumference, which are median and interquartile range
Maternal characteristics N (%) or median (IQR)
Maternal characteristics Age (n=38) 29.5 (26.5-34)
BMI (n=34) 22.5 (19.4-25.6)
Smoking (n=37)
>5 cigarettes/day 7 (19%)
<=5 cigarettes/day 4 (11%)
Alcohol (n=36)
<=1 drink/day 2 (6%)
>1 drink/day 1 (3%)
GW at first contact (n=38) 7 (5.6-11.1)
Previous pregnancies (n=38)
0 17(45%)
1 8 (21%)
2 13 (34%)
Pregnancy outcome and neonatal characteristics N (%) or median (IQR)
Pregnancy outcome (n=38) Spontaneous abortion 3
ETOP 3
Live birth 32
Pregnancy complications (n=32, only live births) Gestational diabetes 5 (16%)
Pre-eclampsia 2 (6%)
Cesarean section 15 (46%)
Preterm birth (<GW 37) 3 (9%)
Neonatal characteristics (n=33, including one pair of twins) GW at birth (n=32) 39 (37.9-39.9)
Weight, g (n=32) 3300 (2935-3610)
Length, cm (n=31) 50 (49-52.5)
HC, cm (n=30) 35 (34-36)
SGA (n=32) 2 (6%)
BMI, Body Mass Index; ETOP, elective termination of pregnancy; GW, gestational week; HC, head circumference; IQR, interquartile range; SGA, small for gestational age (<10th percentile)
Ivabradine exposure and co-medication
The most prevalent indication for ivabradine treatment was supraventricular tachycardia (n = 29) followed by heart failure (n = 3) and coronary artery disease (n = 2). Other reported indications (n = 4) included QT prolongation and sick-sinus-syndrome, although both are regarded as contraindications for ivabradine. The median daily ivabradine dose was 7.5 mg (IQR 5–10, min–max 2.5–15, n = 33).
Nearly all women were treated with ivabradine during the first trimester (n = 37) and 32 stopped treatment after recognition of pregnancy (Fig. 3), at median gestational week 6.1 (interquartile range 5.4–7.8). Cardiac decompensation or complications were not described after cessation of treatment. Only 5 women continued ivabradine throughout pregnancy or until shortly before delivery. One woman started treatment after first trimester (GW 18) of her twin pregnancy due to symptomatic tachycardia under beta-blocker therapy (Iv34 in Fig. 3).Fig. 3 Ivabradine exposure and pregnancy outcome of prospectively (n = 38) and retrospectively (n = 3) ascertained pregnancies
Cardiovascular co-medication consisted of beta-blockers (mainly metoprolol) in 15 women, mostly concomitantly (n = 7) or in replacement of ivabradine (n = 6). Two women initiated ivabradine after discontinuation of beta-blockers (Iv30 and Iv34 in Fig. 3). Six women received other cardiovascular medication, three methyldopa or amlodipine for hypertension, two digitoxin or verapamil for tachyarrhythmia, and one sildenafil and eplerenone for cardiomyopathy with cardiac failure (Iv17).
Psychiatric co-medication was reported in 7/38 women (5 of these with an anxiety disorder) including the teratogenic antiepileptic topiramate (n = 1) and pregabalin (n = 3).
Pregnancy outcome
Of the 38 prospectively ascertained ivabradine-exposed pregnancies, 32 were live births (33 infants including one pair of twins) and three resulted in early spontaneous abortions (Iv10, Iv17, and Iv19 in Fig. 3). One of these occurred in a woman with cardiomyopathy (Iv17) and various co-medications including hydroxycarbamide for thrombocythemia until GW 7. Three pregnancies were electively terminated for social/psychological reasons. Further information on pregnancy complications and outcomes is given in Table 1 and Fig. 3.
Major birth defects were reported in one pregnancy (Iv24) with 15 mg ivabradine plus bisoprolol until GW 7 + 5, both replaced by metoprolol. Polyhydramnios and preeclampsia led to Caesarean section in GW 35 after premature rupture of membranes. The premature child was large for gestational age (> 97 percentile) and had a cleft palate, an atrial septal defect II (spontaneously closed at the age of 10 months), and a hypertrophic cardiac septum. The mother was treated for sinus tachycardia, in addition she had diabetes type I and a BMI of 30.
Neonatal characteristics
Characteristics of the 33 neonates are summarized in Table 1. Compared to the German perinatal survey, standardized birth weights were lower with a median SDS of − 0.36, corresponding to a reduction of birth weight of 149 g for singletons born in gestational week 40.
Two neonates were small for gestational age, one of the twins (exposure only during 2nd and 3rd trimester, Iv34) and one after maternal ivabradine exposure until GW 12 (Iv29). Two neonates were large for gestational age, both mothers were diabetic (Iv24, type I diabetes and Iv27, gestational diabetes).
Retrospective cases
During the study period three pregnancies were retrospectively reported (further details are given in Table 2 and Fig. 3). One premature neonate (IvR1) with tracheal atresia died perinatally. The mother had been treated with ivabradine, naproxen, and metamizole during the first trimester until recognition of pregnancy (no further details available).Table 2 Retrospectively reported pregnancies
Case Maternal age/BMI Treatment indication Ivabradine exposure, other co-medication (GW or trimester) Pregnancy outcome, GW, birth weight percentile, sex Pregnancy complications, birth defects, and additional findings
IvR1 31 years/31.2 Tachycardia Ivabradine (1st trim)
Naproxen (1st trim)
Metamizole (1st trim)
Live birth, 36 + 2, > 97th, f Polyhydramnios, perinatal death, tracheal atresia, normal karyotype (46, XX)
IvR2 29 years/18.4 Tachycardia, syncopes, pacemaker Ivabradine (1st trim)
Bisoprolol (0–35 + 5)
Levetiracetam (0–35 + 5, grand mal epilepsy)
Live birth, 35 + 5, < 10th, f IUGR (since GW 30), oligohydramnios (since GW 34), Caesarean section (vaginal bleeding)
IvR3 29 years/not available Tachycardia Ivabradine (0–41 + 3)
Bisoprolol (0–41 + 3)
Live birth, 41 + 3
> 10th, m
Uncomplicated pregnancy
BMI, body mass index; f, female; GW, gestational week; IUGR, intrauterine growth restriction; m, male; trim, trimester
The other two retrospective cases include a woman with comorbidities, intrauterine growth restriction, and premature birth (IvR2). The second was a healthy newborn (IvR3) after treatment with ivabradine throughout pregnancy. This mother reported a previous also uneventful pregnancy course under ivabradine.
Discussion
Pregnancy outcome
In our prospective cohort, there was only one neonate with major birth defects (cleft palate and atrial septal defect) from a diabetic mother with supraventricular tachycardia. No further major or cardiac malformations were reported among the 32 live births and 6 pregnancy losses. Cardiac defects have been observed in preclinical experimental studies. As a possible teratogenic mechanism, ivabradine-induced embryonic heart rate reduction leading to reduced cardiac output and hypoxia was discussed [16]. With a prevalence of almost 1 of 100 cardiac malformations belong to the most frequent birth defects [17, 18]. In our case, the neonate was large for gestational age and had a septum hypertrophy, supporting the assumption that maternal pre-gestational diabetes may also be causative for the observed cardiac defect and cleft palate. A causal relationship between ivabradine and cleft palate is unlikely, because ivabradine was discontinued at GW 7 + 5 before the beginning of the critical time period for the development of cleft palate at gestational week 9. The rate of major birth defects in our prospective cohort is comparable to the expected background rate of 2–3% [17, 18]. However, the limited number of pregnancies exposed during the entire period of embryogenesis does not allow excluding an increased risk, so far.
There was only one retrospectively reported case with a birth defect (malformation of the trachea) suspected as potentially ivabradine related. The concomitant exposures with naproxen and metamizole have not been evidenced as teratogens in human, whereas the high maternal BMI in the present case poses a risk for birth defects.
Taken together, we did not observe a specific pattern of congenital anomalies in our study. In addition, the presence of maternal comorbidities like diabetes and high BMI may have contributed to the observed birth defects.
The number of spontaneous abortions (3/38) was not higher than expected. The rate of preeclampsia and preterm birth in our cohort was within the expected range [19].
Treatment indications
Although the approved treatment indications for ivabradine, chronic heart failure and chronic stable angina pectoris (see Fig. 2 for changes of licenced use during the study period), are rare in women of reproductive age, we have observed an increasing need for information for off-label use in (younger) women considering pregnancy.
In our case series, most women were treated (off-label) for supraventricular tachycardia using lower doses than recommended (5–7.5 mg twice daily). Supraventricular tachycardia is not uncommon, especially in younger women, and is expected in 0.02–0.5% of pregnancies [20]. Aggravation of symptoms during pregnancy is described in about 20–50% of women related to various mechanisms, including pregnancy-related increase in cardiac heart rate and output and hormonal changes [20].
In 2015, the HRS (Heart Rhythm Society) considered ivabradine as promising for treating patients with inappropriate sinus tachycardia (IST) [13], followed by a recommendation for ivabradine by the ACC (American College of Cardiology) in 2016 [14]—though only one small randomized controlled study has yet supported beneficial effects [21]. Although still not approved for this indication, recent ESC guidelines [15] also recommend ivabradine alone or in combination with a beta-blocker in symptomatic patients with IST, for the treatment of focal atrial tachycardia or for postural orthostatic tachycardia syndrome (POTS), a condition considered as common cause for orthostatic intolerance in women between 15 and 25 years [20, 22] (Fig. 2).
Beta-blockers were required in 18% of women in our prospective cohort (Fig. 3). Discontinuation of ivabradine or switching to beta-blockers after recognition of pregnancy was well tolerated in most of our patients. Thus, necessity of ivabradine treatment should be critically considered in women with supraventricular tachycardia planning pregnancy. Data from previous studies suggest that discontinuation of ivabradine is possible in many (non-pregnant) patients with IST without recurrence. Even beneficial long-term effects after cessation of therapy are discussed [23, 24].
A high number of women in our cohort received therapy for anxiety and/or other psychiatric conditions. This is in line with findings from other studies reporting high rates of anxiety (25%) and depression (26%) in IST patients [25]. Anxiety is considered as non-cardiac symptom of inappropriate sinus tachycardia, and vice versa, anxiety may trigger tachycardia. Especially during pregnancy, empathic care and reassurance are necessary, to prevent increased concerns about possible harm to the fetus [26].
Preconception counselling is of utmost importance in all women considering pregnancy under medication [27]. Sinus tachycardia is a condition diagnosed mainly in young females and the first diagnosis is frequently made during or after pregnancy [25]. Considering the increasing off-label use of ivabradine as well as the high rate of > 40% unplanned pregnancies [28], more exposed pregnancies can be expected in the future. Well-established treatment options such as selected beta-blockers with low reproductive risk [29] should be considered in women of childbearing age before prescription of ivabradine.
Strength and limitations
A major strength of our study is the detailed exposure protocol and the multi-source ascertainment of pregnancy outcome via patients and their HCP followed by a case by case plausibility test.
Our data may not be representative of the German pregnant population; insofar, women with higher education are over-represented [30]. However, the regional distribution of enrolled patients is representative of the female population of childbearing age in Germany [11]. In addition, women on long-term ivabradine with severe cardiovascular conditions such as the approved treatment indications coronary artery disease and cardiac failure are under-represented in our cohort possibly favouring uneventful pregnancy outcomes. However, given the rarity of severe cardiac conditions in pregnant women, large study cohorts are not to be expected in the foreseeable future.
Conclusions
Our findings do not indicate that ivabradine is a major teratogen when used in early pregnancy. Although being the largest study published so far on the outcome of ivabradine-exposed pregnancies, the sample is too small to rule out embryotoxic effects. Further studies are needed to confirm or refute our findings.
In women planning pregnancy, established drugs of choice with low risk for the unborn should be preferred to ivabradine. For women with significant arrhythmias and severe cardiac disease, the impact of the underlying medical condition including comorbidities such as diabetes and high BMI on pregnancy outcome has to be considered when planning pregnancy.
After inadvertent ivabradine exposure during pregnancy or lack of treatment alternatives, fetal ultrasound for structural anomalies and growth restriction is recommended. In addition, all pregnant women with complicated arrhythmias or major cardiac disease should be carefully monitored.
Acknowledgements
We would like to thank all patients and physicians reporting pregnancies to the institute and acknowledge our colleagues from the German Embryotox Institute for counselling and thoroughly documenting cases, especially Sandra Zinke for continuous support with case documentation.
Contributions of authors
MH, ML, and CS designed the study. ML, EB, MH, and KD validated and analysed the data. ML, MH, and CS wrote the first draft of the manuscript, and all the authors critically revised subsequent manuscript drafts and contributed essential discussion points. All the authors approved the final version of this manuscript and they are responsible for the accuracy of this work.
Funding
Open Access funding enabled and organized by Projekt DEAL. This work was performed with financial support from the German Ministry of Health (BMG) and the German Federal Institute for Drugs and Medical Devices (BfArM). The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
Ethical approval
The study was registered with the German Clinical Trial register (DRKS00010502) and the study protocol was approved by the ethics committee of the Charité-Universitätsmedizin Berlin (EA1/107/16).
The study was registered with the German Clinical Trial register (DRKS00010502, 20.5.2016)
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. | Recovered | ReactionOutcome | CC BY | 33501507 | 19,157,674 | 2021-07 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.